Skip to main content
NCBI home page
Journal of Anatomy logoLink to Journal of Anatomy
. 2017 Jul 18;231(4):585–599. doi: 10.1111/joa.12649

Homology, homoplasy and cusp variability at the enamel–dentine junction of hominoid molars

Alejandra Ortiz 1,✉, Shara E Bailey 2,3, Jean‐Jacques Hublin 3, Matthew M Skinner 3,4
PMCID: PMC5603786  PMID: 28718921

Abstract

Evolutionary studies of mammalian teeth have generally concentrated on the adaptive and functional significance of dental features, whereas the role of development on phenotypic generation and as a source of variation has received comparatively little attention. The present study combines an evolutionary biological framework with state‐of‐the‐art imaging techniques to examine the developmental basis of variation of accessory cusps. Scholars have long used the position and relatedness of cusps to other crown structures as a criterion for differentiating between developmentally homologous and homoplastic features, which can be evaluated with greater accuracy at the enamel–dentine junction (EDJ). Following this approach, we collected digital models of the EDJ and outer enamel surface of more than 1000 hominoid teeth to examine whether cusp 5 of the upper molars (UM C5) and cusps 6 and 7 of the lower molars (LM C6 and LM C7) were associated each with a common developmental origin across species. Results revealed that each of these cusps can develop in a variety of ways, in association with different dental tissues (i.e. oral epithelium, enamel matrix) and dental structures (i.e. from different cusps, crests and cingula). Both within and between species variability in cusp origin was highest in UM C5, followed by LM C7, and finally LM C6. The lack of any species‐specific patterns suggests that accessory cusps in hominoids are developmentally homoplastic and that they may not be useful for identifying phylogenetic homology. An important and unanticipated finding of this study was the identification of a new taxonomically informative feature at the EDJ of the upper molars, namely the post‐paracone tubercle (PPT). We found that the PPT was nearly ubiquitous in H. neanderthalensis and the small sample of Middle Pleistocene African and European humans (MPAE) examined, differing significantly from the low frequencies observed in all other hominoids, including Pleistocene and recent H. sapiens. We emphasize the utility of the EDJ for human evolutionary studies and demonstrate how features that look similar at the external surface may be the product of different developmental patterns. This study also highlights the importance of incorporating both developmental and morphological data into evolutionary studies in order to gain a better understanding of the evolutionary significance of dental and skeletal features.

Keywords: accessory cusps, dental development, developmental homoplasy, hominoid molars

Introduction

The concept of homology has been the subject of intense debate since it was delineated in the mid‐19th century by Sir Richard Owen as ‘the same organ in different animals under every variety of form and function’. The history of modern debates on homology has been extensively documented elsewhere (Patterson, 1982; Hall, 1994). Within an explicitly phylogenetic framework, homology is defined as similarity between taxa that is inherited from their last common ancestor (LCA), and is distinguished from homoplasy, which is regarded as any morphological resemblance that results from processes other than common ancestry (Simpson, 1961; Hennig, 1966; Patterson, 1982; Lieberman, 1999; Lockwood & Fleagle, 1999). Unless otherwise noted, we use these definitions here.

Because the fossil record of many mammals is represented predominantly by teeth, studies on cusp homologies have been central for understanding the evolution of tribosphenic molars from single‐cusped reptilian teeth, and concomitantly early mammalian evolutionary history (Butler, 1939, 1978, 1990; Patterson, 1956; Hershkovitz, 1971). These studies also document the limited number of ways in which teeth can vary and evolve. All therian mammals form a monophyletic group, which descended from a common ancestor with tribosphenic molars in which the paracone, protocone and metacone of the upper molars form the trigon, and the protoconid, paraconid and metaconid of the lower molars form the trigonid (Hershkovitz, 1971; Luo et al. 2001). However, accessory cusps and cuspules forming in addition to these structures are particularly prone to parallel evolution. In fact, considering the great diversity of mammalian tooth shapes, homoplasy of dental features appears to be pervasive (Jernvall, 1995). Butler (1978), for example, reported that the mesostyle and metastylid have developed independently in several groups, and noted that the exact developmental origin of these and other structures is generally unknown. In a classic example, Hunter & Jernvall (1995) noted that the hypocone has independently evolved more than 20 times in mammals (see also Butler, 1956; Van Valen, 1982). They found that this cusp most commonly derived from either the lingual cingulum or the metaconule. The hypocone may also develop from the metacone or protocone, and in some taxa its mode of origin remains unknown. Even within the primate order the hypocone may have evolved more than once (Gregory, 1922; Butler, 2000; but see Butler, 1963; Hershkovitz, 1977). This appears to be the case for Eocene primates whose hypocone evolved from the lingual cingulum in European adapines and from the Nannopithex fold in North American notharctines (Gregory, 1922; Butler, 2000; Anemone et al. 2012).

Given the high likelihood for homoplasy, Butler (1978, 1985) and Van Valen (1994) proposed that cusp names should only be used as topographical terms, without implying phylogeny. However, even the ‘identity’ of cusps has sometimes proven difficult to determine. For example, Sánchez‐Villagra & Kay (1996) disproved the long‐held view that the upper molars of diprotodont marsupials possessed a metaconule rather than a hypocone. More recently, Jernvall et al. (2008) found that the paracone of Hapalemur simus upper premolars has shifted distally to become the metacone. Jernvall et al. (2008) also stress the need to incorporate developmental data in identifying homoplastic features from ‘true’ similarity. Direct experimental testing in fossils is impossible, and in most extant mammals it is unfeasible. However, addressing cusp homology/homoplasy within a developmental framework can be at least partially achieved through the analysis of the internal surface of teeth at the enamel–dentine junction (EDJ), as shown by the pioneering studies of Kraus (1952), Korenhof (1960), Corruccini (1987), and Corruccini & Holt (1989).

The EDJ is the interface between the enamel cap and dentine crown, and preserves the end point of growth of the inner enamel epithelium, whose size and shape determine the main crown configuration (Schour & Massler, 1940; Butler, 1956; see also Skinner, 2008; Ortiz et al. 2012; Morita et al. 2016). Recently, Anemone et al.'s (2012) study of adapid upper molars at the EDJ supported early assessments by Gregory (1922) and Butler (2000) that the hypocone evolved convergently among closely related primate groups from the Eocene. In agreement with Guatelli‐Steinberg & Irish's (2005) suggestion that cusp 6 of the lower molars may be homoplastic within the hominin lineage, a preliminary study of the EDJ by Skinner et al. (2008) documented that in hominoids this cusp can form in association with the hypoconulid or within the distal fovea. They also found that cusp 7 can originate in two developmentally different ways, such that it can derive from either the metaconid or interconulid.

The presence and degree of expression of accessory cusps have been used widely in species diagnoses and phylogenetic reconstructions of the hominin fossil record (Wood et al. 1983; Wood & Engleman, 1988; Suwa et al. 1996; Bailey, 2002, 2004; Bailey & Wood, 2007; Martinón‐Torres et al. 2007, 2008, 2012; Bailey et al. 2009; Irish et al. 2013). However, it is unknown whether or not each of these cusps is associated with a single developmental origin and, therefore, the evolutionary implications of these dental features remain uncertain. Building upon previous studies by Skinner et al. (2008, 2014) and Anemone et al. (2012), we use micro‐computed tomography (microCT) to assess accessory cusp variation at the EDJ in a taxonomically broad sample of 1168 extant and fossil hominoid molars. Specifically, we focus our analyses on cusp 5 of the upper molars (also known as the metaconule by Turner et al. 1991; but see below) and cusps 6 and 7 [Turner et al.'s (1991) entoconulid/tuberculum sextum and metaconulid/tuberculum intermedium, respectively] of the lower molars. We examine the different developmental ways in which these accessory cusps and cuspules can form in the hominoid lineage and whether or not there are any species‐specific patterns that can inform us about homology and homoplasy. To that end, an assumption of the study is that if an accessory cusp is preferentially located on a dental structure (relative to other structures within a tooth), the two features are developmentally linked (Skinner et al. 2008, 2014; Anemone et al. 2012; see also Van Valen, 1994; Hunter & Jernvall, 1995). Our main hypothesis is that accessory cusps and cuspules that look superficially similar at the external surface may have different developmental origins, and that the likelihood of homoplasy of these accessory cusps and cuspules in hominoids is high. We also examine variation in the post‐paracone tubercle (PPT) of the upper molars and its usefulness for hominoid systematics.

Materials and methods

Study sample

Our sample includes three‐dimensional (3D) models of the EDJ and outer enamel surface (OES) of 466 upper and 702 lower molars of extant and extinct hominoids (Table 1). All data derive from original specimens subjected to microCT. The fossil sample comprises the following species (with number of teeth in brackets): Australopithecus anamensis (n = 17), A. afarensis (n = 20), A. africanus (n = 112), Paranthropus aethiopicus (n = 2), P. boisei (n = 14), P. robustus (n = 113), Homo sp. (n = 21; mainly specimens attributed to H. habilis sensu lato), H. erectus s. l. (n = 14), the Middle Pleistocene African and European group (MPAE; n = 8), H. neanderthalensis (n = 147), and Pleistocene H. sapiens (n = 66). This reflects the taxonomic nomenclature most commonly used by researchers to date. The inclusive categories H. habilis s.l. H. erectus s.l. and MPAE were used here given the small number of available specimens assigned to these groups. The detailed list of the fossil specimens used can be found in Tables S1 and S2.

Table 1.

Sample composition for hominoid upper and lower molars used in this study

Taxon UM1 UM2 UM3 UM Total UM LM1 LM2 LM3 LM Total LM Total
A. anamensis 3 3 0 0 6 4 4 3 0 11 17
A. afarensis 3 1 3 1 8 4 6 2 0 12 20
A. africanus 13 18 19 1 51 18 19 24 0 61 112
P. aethiopicus 0 0 0 0 0 0 2 0 0 2 2
P. boisei 2 2 2 0 6 0 4 2 2 8 14
P. robustus 17 14 17 1 49 20 20 24 0 64 113
H. habilis s.l. 6 3 2 0 11 3 3 1 3 10 21
H. erectus s.l. 2 3 1 1 7 2 3 2 0 7 14
MPAE 2 1 2 0 5 1 1 1 0 3 8
H. neanderthalensis 21 25 19 1 66 33 27 20 1 81 147
H. sapiens (Pleistocene) 10 8 3 2 23 10 17 11 5 43 66
H. sapiens (recent) 18 41 14 13 86 47 86 41 7 181 267
P. troglodytes 23 29 18 0 70 42 51 16 0 109 179
P. paniscus 5 3 0 0 8 12 14 0 0 26 34
Gorilla sp. 11 13 12 0 36 10 13 13 0 36 72
Pongo sp. 14 11 9 0 34 20 19 9 0 48 82
Total per tooth type 150 175 121 20 466 226 289 169 18 702 1168
Open in a new tab

MPAE, Middle Pleistocene African and European humans.

Extant samples include contemporary H. sapiens (n = 267), Pan paniscus (n = 34), P. troglodytes ssp. (n = 179), Gorilla sp. (n = 72) and Pongo sp. (n = 82). The contemporary H. sapiens sample comprises individuals of European or African ancestry, or of unknown geographic provenience. The P. troglodytes sample includes P. t. verus (UM = 39 and LM = 54), P. t. troglodytes (UM = 11 and LM = 23), P. t. schweinfurthii (UM = 8 and LM = 7) and P. troglodytes of unknown subspecific affiliation (UM = 12 and LM = 25). The Gorilla sample includes G. gorilla (UM = 34 and LM = 31) and G. beringei (UM = 2 and LM = 5), and that of Pongo consists of P. pygmaeus (UM = 8 and LM = 15), P. abelii (UM = 14 and LM = 12) and Pongo sp. (UM = 12 and LM = 21). Sample size per trait varies due to differential preservation and wear. Although we did not include known antimeres, some individuals are represented by more than one molar (Tables S1 and S2). Given that sex is unknown for most fossil specimens, we made no attempt to control for sex. However, it has been demonstrated that, with few exceptions, dental morphological traits show no consistent sexual dimorphism in living humans (Scott & Turner, 1997). This is also true for extant great apes, despite marked differences in tooth size between males and females (Uchida, 1996; Pilbrow, 2003).

Data collection procedures and analyses

Each specimen was scanned using microCT, with either a BIR ACTIS 225/300 (130 kV, 100 μA, 0.25 brass filter) or a Skyscan 1172 (100 kV, 94 μA, 2.0 mm aluminum and copper) scanner. Pixel dimensions and slice spacing of the resultant images ranged between 10 and 30 microns. The complete image stack of each tooth was filtered using a computer‐programmed macro that employs a 3D median and mean‐of‐least‐variance filter (each with a kernel size of one or three) to improve tissue gray‐scale homogeneity and facilitate tissue segmentation (Wollny et al. 2013). Filtered image stacks were imported into the Avizo (FEI Visualization Sciences Group), and enamel and dentine tissues were segmented manually. Only teeth with well‐distinguished gray‐scale pixel values and thus with a clear separation of the enamel and dentine tissues were segmented. Digital surface models (.ply format) of the EDJ and OES were produced in Avizo using the surface generation module with the unconstrained smoothing parameter.

The definition of cusp 5 of the upper molars (UM C5) and cusps 6 and 7 of the lower molars (LM C6 and LM C7, respectively) followed standards outlined by the Arizona State University Dental Anthropology System (ASUDAS; Turner et al. 1991). In order to assess the degree of trait correspondence between the EDJ and OES, cusps were classified as present at a given surface if any expression other than ASUDAS grade 0 was detected. When more than one cusps/cuspules were present at the ‘normal’ position of a given accessory cusp, these were considered (at least preliminarily) part of the same accessory cusp complex [e.g. LM ‘double’ C6 reported by Bailey & Wood (2007) and Skinner et al. (2008)]. For each trait, the correspondence between the EDJ and OES was examined using the following scoring system: (i) grade 0: accessory cusp absent; (ii) grade 1: one cusp present; (iii) grade 2: two cusps present; and (iv) grade 3: three or more cusps present. This system allowed us to assess whether these cusps can develop entirely from enamel deposition. Following Skinner & Gunz (2010), a ‘suspected’ category was also included to incorporate those cases where it was unclear whether or not an accessory cusp was present [see also Turner et al. (1991) for additional examples of indecisive categories]. Each ‘suspected’ accessory cusp was given a score of 0.5. Data were collected both at the EDJ and OES and, to avoid errors associated with worn or poorly preserved teeth, only complete molars with little‐to‐no dental wear [equivalent to Molnar's (1971) first three wear stages] were included for analysis. Concordance in trait expression between the two surfaces was tested using the non‐parametric Spearman's rank correlation coefficient, calculated in PAST (Hammer et al. 2001).

If the accessory cusp was not entirely the result of enamel deposition, its origin was examined at the EDJ. Following Van Valen (1994) and Hunter & Jernvall (1995), assessments of trait origin were based on topological relationships between two given dental structures, such as crests, cusps/cuspules or cingula. For this purpose, tooth nomenclature followed Szalay (1969), Rosenberger & Kinzey (1976) and Swindler (2002). We acknowledge, however, that while the presence of different topological associations is suggestive of homoplasy, the sharing of similar dental associations between taxa does not necessarily indicate homology nor does it directly reflect phylogeny, as accessory cusps linked by the same developmental process could still undergo convergence. As they represent the majority of cases, analyses of trait origin focused on single‐cusped features, although data on their multi‐cusped variants are also briefly discussed. All teeth were scored twice, with scoring sessions separated by at least one month. When discrepancies between the two scoring sessions occurred, trait presence and origin were scored a third time and scores that matched between two given assessments were used as final data points. This third scoring session was also separated by one month from the second one. All molar types were pooled into two categories (upper and lower molars) to maximize sample sizes per taxon.

We also evaluated the PPT of the upper molars, which occurs on the distal slope of the paracone. This feature was identified in H. neanderthalensis by Martin et al. (2017), but until now its presence and variation in other extant and fossil hominoids has not been assessed. This trait should not be confused with the lingual paracone tubercle, which occurs on the occlusal surface, distal to the mesial marginal tubercles (Kanazawa et al. 1990). Expressions of the PPT were classified into four categories: (i) grade 0: PPT absent (distal slope of paracone is smooth); (ii) grade 1: shouldering present only; (iii) grade 2: faint‐to‐moderate tubercle present; and (iv) grade 3: marked tubercle present (Fig. 1). For PPT, the significance of the observed patterns was tested via bootstrapping (1000 iterations) performed in R (R Core Team, 2012).

Figure 1.

Figure 1

Open in a new tab

Three‐dimensional models of the enamel–dentine junction (EDJ) of four upper molars illustrating variation in post‐paracone tubercle (PPT) expression. (A) Grade 0: PPT absent (extant H. sapiens depicted); (B) grade 1: shouldering present (H. neanderthalensis KRP_D169 depicted); (C) grade 2: faint‐to‐moderate tubercle present (H. neanderthalensis KRP_D96 depicted); and (D) grade 3: marked tubercle present (H. neanderthalensis Scladina 4A_4 depicted).

Results

Cusp 5 of the upper molars

Results of the Spearman's correlation coefficient provided in Table 2 reveal a high and significant concordance between UM C5 expressions at the EDJ and OES in extant great apes, H. sapiens (Pleistocene and recent) and H. neanderthalensis. For these taxa, only subtle differences in trait expression at the EDJ and OES were observed, with correlation coefficients ranging between 0.829 and 1. With the exception of two H. neanderthalensis specimens (see below), discrepancies always involved the ‘suspected’ category. These discrepancies in most cases occur when subtle or blunt dentine horns were classified as ‘suspected’ at the EDJ, but UM C5 was either present or absent at the OES. Interestingly, two molars from a sample of 41 Neandertal specimens exhibited a UM C5 cuspule at the OES with no equivalent structure at the EDJ. Although sample sizes for P. paniscus and MPAE were too small to run any statistical analyses, no trait expression differences were observed in the specimens examined.

Table 2.

Spearman's correlation coefficient test for UM C5 expression at the EDJ and OES

Taxon n r P‐value
East African Australopithecus a 14 0.662 < 0.01
A. africanus 38 0.638 < 0.001
P. robustus 32 0.560 < 0.001
Early Homo b 10 0.697 < 0.05
H. neanderthalensis 41 0.857 < 0.001
Pleistocene H. sapiens 9 0.922 < 0.01
Recent H. sapiens 66 0.833 < 0.001
P. troglodytes 54 0.829 < 0.001
Gorilla sp. 32 0.905 < 0.001
Pongo sp. 24 1.000 < 0.001
Open in a new tab

Significant values in bold.

a

Includes A. anamensis and A. afarensis.

b

Includes H. habilis s.l. and H. erectus s.l.

The concordance between UM C5 expression at the EDJ and OES for Australopithecus, Paranthropus and early Homo species was moderate, with correlation coefficients ranging from 0.56 to 0.697 (Table 2). All correlations were statistically significant, with P. robustus showing the lowest correlation between the two surfaces. The main source of discrepancy in these three groups was the result of one or more UM C5 cuspules present at the OES with no associated dentine horn(s) on its underlying surface. Although such cases were primarily represented by specimens with no UM C5 at the EDJ and one cuspule at the OES, there were instances in which UM C5 was present at both surfaces but the number of dentine horns at the EDJ did not correspond to the number of cuspules at the OES (Fig. 2). This was particularly evident in Paranthropus. It should be noted, however, that in no case was a moderate‐sized or large UM C5 observed as present at the external surface when a dentine horn was absent at the EDJ.

Figure 2.

Figure 2

Open in a new tab

Paranthropus robustus (SK 831a ULM3 mirror‐imaged) showing one dentine horn associated with UM C5 at the enamel–dentine junction (EDJ) (A), but three cuspules at the outer enamel surface (OES) (B). The white arrows indicate the presence and location of the dentine horn and cuspules at the EDJ and OES, respectively. Distal to the left.

Table 3 summarizes UM C5 frequency by taxon and developmental origin. Although the frequency of occurrence of this cusp was low in most taxa, when present, UM C5 may have its origin on the hypocone, metacone or distal fovea/middle portion of the distal marginal ridge (Fig. 3). It may also arise directly from the buccal cingulum or from the occlusal surface in association with a distal crest (e.g. crest connecting the distal ridge with either the metacone or hypocone, or an independent crest on the distal fovea). Weak expressions of UM C5 can also appear later during tooth morphogenesis as a result of enamel deposition only. Although UM C5 most frequently arises as an outgrowth of the distal fovea/middle portion of the distal marginal ridge, there is a high degree of variability in its origin, both within and between species (Table 3). A. africanus exhibits a unique pattern in which UM C5 originates from the buccal cingulum in the majority of cases (42.9%). This is followed by cases of UM C5 derived from enamel deposition only (28.6%) and from the distal fovea (19%). The origin of UM C5 at the buccal cingulum was not observed in other taxa (but see Fig. 3). Thick‐enameled and megadont P. robustus and P. boisei also exhibited a relatively high frequency of UM C5 derived entirely from enamel deposition (39.1% and 33.3%, respectively), although in the majority of cases this cusp appears earlier during tooth development at the EDJ and arises from the distal fovea. In all cases of enamel‐derived UM C5s, the cuspule does not exceed ASUDAS grade 2. Chimpanzees also show a distinct pattern in which UM C5 originates from the hypocone in the highest frequency (57.1%). The high frequency of UM C5 deriving from the hypocone in P. troglodytes contrasts with the low incidence (0–15.4%) of this variant in other taxa.

Table 3.

Variation in origin of UM C5 per taxon

Taxon Total Absent, % Present, n (%) Hypocone, % Metacone, % Distal fovea, % Buccal cingulum, % Occlusal surface, % Enamel, %
A. anamensis 5 100.0 0 (0.0) – – – – – –
A. afarensis 8 50.0 4 (50.0) 0.0 0.0 75.0 0.0 0.0 25.0
A. africanus 43 51.2 21 (48.8) 0.0 4.8 19.0 42.9 4.8 28.6
P. boisei 5 40.0 3 (60) 0.0 0.0 66.7 0.0 0.0 33.3
P. robustus 45 48.9 23 (51.1) 0.0 13.0 47.8 0.0 0.0 39.1
Homo sp./habilis s.l. 12 75.0 3 (25.0) 0.0 33.3 66.7 0.0 0.0 0.0
H. erectus s.l. 7 85.7 1 (14.3) – – 100.0 – – –
MPAE 4 25.0 3 (75.0) – 33.3 66.7 – – –
H. neanderthalensis 46 71.7 13 (28.3) 15.4 0.0 69.2 0.0 0.0 15.4
H. sapiens (Pleistocene) 17 41.2 10 (58.8) 10.0 0.0 90.0 0.0 0.0 0.0
H. sapiens (recent) 78 75.6 19 (24.4) 5.3 0.0 94.7 0.0 0.0 0.0
P. troglodytes 63 66.7 21 (33.3) 57.1 9.5 28.6 0.0 4.8 0.0
P. paniscus 6 100.0 0 (0.0) – – – – – –
Gorilla sp. 33 78.8 7 (21.2) 14.3 14.3 71.4 0.0 0.0 0.0
Pongo sp. 25 92.0 2 (8.0) 0.0 0.0 100.0 0.0 0.0 0.0
Open in a new tab

MPAE, Middle Pleistocene African and European humans.

Frequencies of origin types per taxon do not include individuals with trait absence.

Figure 3.

Figure 3

Open in a new tab

Types of UM C5 development. (A) UM C5 absent [enamel–dentine junction (EDJ) of P. t. verus illustrated]; (B) hypocone type (EDJ of P. t. troglodytes illustrated); (C) metacone type (EDJ of A. africanus MLD 28 URM3 illustrated with close‐up of distobuccal view inset in bottom left corner); (D) distal fovea type (EDJ of P. robustus SKX 21841 URM3 illustrated); (E) buccal cingulum type (EDJ of A. africanus Sts 28 URM2 illustrated with close‐up of distobuccal view inset in bottom left corner); (F) occlusal type (EDJ of A. africanus Sts 52a URM3 illustrated); and (G and H) enamel type [EDJ and outer enamel surface (OES) of P. robustus SK 13.14 URM2 illustrated]. We followed a conservative approach and differentiated UM C5s derived from the distal slope of the metacone from those arising from the buccal cingulum as they appear to be morphologically different. However, this could be the result of remnants of the buccal cingulum being more common and/or more strongly expressed in some taxa such as A. africanus.

When present, UM C5 most commonly occurs as a single‐cusped feature. However, cases of two or more ‘UM C5’ dentine horns observed at the EDJ were also found in most groups (Table S3). In these cases, the dentine horns may have their origins from the same (e.g. hypocone) or different (e.g. hypocone and buccal cingulum) structures. Most cases of multiple ‘UM C5’ dentine horns involved the distal fovea/distal marginal ridge. Among samples with more than 10 observations, the presence of two ‘UM C5’ cusps deriving from two different structures was highest in A. africanus and H. neanderthalensis (8.3% and 7.0%, respectively).

PPT of the upper molars

Frequencies and degrees of expression of the PPT at the EDJ are provided in Table 4. Results reveal that this feature is nearly ubiquitous in Neandertals, with only 1.6% of the 64 specimens examined showing a smooth surface on the distal paracone. When present, 70.3% of Neandertal upper molars exhibit either a pronounced or blunt additional dentine horn, distal to that associated with the tip of the paracone. The remaining Neandertal specimens examined (28.1%) exhibit at least some shouldering on the distal slope of this cusp. Steinheim (UM1, UM2 and UM3) and Thomas Quarry I (UM1 and UM3) were the only MPAE upper molars available for study. Both show some expression of PPT on all molars. The nearly ubiquitous presence of the PPT in H. neanderthalensis (and MPAE if the above sample is representative) contrasts with the low frequency of this trait in all other hominoids, where more than 67% of individuals in each taxon (except for Pleistocene H. sapiens) lack it completely and, when present, the PPT is mainly represented by the shouldering type (Fig. 4). The majority of the Pleistocene H. sapiens teeth lack a PPT (52.4%), but a marked or blunt dentine horn on the distal paracone was found in 28.5% of the sample (9.5% for marked and 19% for blunt PPT). Although most similar to H. neanderthalensis (and MPAE) frequencies compared with other taxa, a value of 28.5% is far below the 70.3% seen among Neandertals. Cases of marked expression of PPT were also found in recent H. sapiens (2.5%), as well as in Pongo (3.6%). The only case seen in Pongo, however, was located more distally on the paracone relative to those observed in Homo species. Table S4 presents the results of the bootstrapping analysis (95% confidence), which reveals that both moderate (blunt) and marked expressions of the PPT in H. neanderthalensis (39.9–59.2% and 17.1–29.6%, respectively) differ significantly from all other hominoid groups examined (0–28.6% for moderate expressions and 0–16.9% for marked expressions), including Pleistocene H. sapiens.

Table 4.

Frequencies of occurrence and expression of the PPT per taxon

Taxon n PPT absent, % PPT shouldering, % PPT faint/moderate, % PPT marked, %
A. anamensis 5 100.0 0.0 0.0 0.0
A. afarensis 7 85.7 14.3 0.0 0.0
A. africanus 46 91.3 8.7 0.0 0.0
P. boisei 6 100.0 0.0 0.0 0.0
P. robustus 43 100.0 0.0 0.0 0.0
Homo sp./habilis s.l. 8 87.5 0.0 12.5 0.0
H. erectus s.l. 6 66.7 33.3 0.0 0.0
MPAE 5 0.0 60.0 40.0 0.0
H. neanderthalensis 64 1.6 28.1 46.9 23.4
H. sapiens (Pleistocene) 21 52.4 19.0 19.0 9.5
H. sapiens (recent) 80 75.0 16.3 6.3 2.5
P. troglodytes 66 89.4 7.6 3.0 0.0
P. paniscus 7 100.0 0.0 0.0 0.0
Gorilla sp. 35 100.0 0.0 0.0 0.0
Pongo sp. 28 85.7 10.7 0.0 3.6
Open in a new tab

MPAE, Middle Pleistocene African and European humans; PPT, post‐parabone tubercle.

Highest frequency per sample in bold.

Figure 4.

Figure 4

Open in a new tab

Buccal view of upper molars with examples of post‐paracone tubercle (PPT) expression on the distal slope of the paracone in H. neanderthalensis (A–F; A: Vi 11–46 M2 B: KMH 21 UM1 C: SD 407 UM1 D: KRP 46 UM1 E: KRP 171 UM1 F: SR 1164 M3) compared with the smooth surface more commonly observed in other hominoids. (G) P. robustus (SK 102 UM1); (H) Pleistocene H. sapiens (Qafzeh 15 UM2); (I) early Holocene H. sapiens (Combe Capelle UM2); (J) Pleistocene H. sapiens (Skhul I UM1); (K) Pleistocene H. sapiens (Qafzeh 9 UM1); and (L) recent H. sapiens. Mesial to the left.

Cusp 6 of the lower molars

Table 5 provides the correlation coefficients for LM C6 expressions at the EDJ and OES. Except for P. robustus (r = 0.63), all taxa show a high correlation in trait expression between the two surfaces, with values ranging between 0.79 and 1. All values are statistically significant (P < 0.001). As in cusp 5 of the upper molars, the few cases of disagreement observed for LM C6 involved the ‘suspected’ category either at the EDJ or OES. These discrepancies resulted primarily from small dentine horns that were not clearly represented by a cusp at the OES. Less frequent were cases in which a LM C6 was ‘suspected’ at the EDJ but either absent or present at the OES. Out of the more than 500 extant and fossil hominoid molars examined, only two specimens (one recent H. sapiens and one P. troglodytes) showed a small but clear dentine horn at the EDJ with no equivalent structure at the OES. In contrast, cases of small LM C6 cuspules produced entirely by enamel deposition were found in P. robustus and P. boisei and, to a lesser extent, in A. africanus. In all cases LM C6 structures resulting entirely from enamel deposition were small (> ASUDAS grade 2).

Table 5.

Spearman's correlation coefficient test for LM C6 expression at the EDJ and OES

Taxon n r P‐value
A. anamensis 11 0.975 < 0.001
A. afarensis 11 0.981 < 0.001
A. africanus 51 0.878 < 0.001
P. robustus 44 0.629 < 0.001
Early Homo a 14 1.000 < 0.001
H. neanderthalensis 35 0.962 < 0.001
Pleistocene H. sapiens 36 1.000 < 0.001
Recent H. sapiens 152 0.941 < 0.001
P. troglodytes 89 0.900 < 0.001
P. paniscus 22 0.788 < 0.001
Gorilla sp. 35 1.000 < 0.001
Pongo sp. 37 1.000 < 0.001
Open in a new tab

Significant values in bold.

a

Includes H. habilis s.l. and H. erectus s.l.

Table 6 summarizes LM C6 trait frequency in developmental origins by taxon as revealed by the examination of 612 lower molars. This study supports Skinner et al.'s (2008, 2014) conclusions suggesting that LM C6 may form in proximity to the dentine horn of either the hypoconulid or entoconid. LM C6 may also arise independently from the distal fovea and, in rare occasions, this cusp may originate from a dentine horn on the entoconid–hypoconulid crest (or an independent crest on the distal fovea) at the distal portion of the occlusal surface (Fig. 5). Cases of enamel‐derived LM C6 with no underlying dentine horn associated with the cusp are rare (2.5–5.9%). Despite the different developmental ways in which LM C6 may form, this cusp appears to be less variable than the UM C5, both within and between species. With some exceptions (see below), all observations from Australopithecus, Paranthropus, H. habilis s.l., MPAE, H. sapiens, P. paniscus and Gorilla suggest that the distal fovea is the primary, and in some species only, source of LM C6 origin. Exceptions include the enamel‐ or occlusal‐derived LM C6s present in A. africanus and P. robustus. The frequency of LM C6 formed on the occlusal surface without involvement of the marginal ridge was also particularly high in Gorilla (25%). Furthermore, H. neanderthalensis is the only hominin sample in which a moderate frequency (24.1%) of LM C6s originated from the hypoconulid. A similar pattern was observed in chimpanzees (31.1%). Finally, cases of LM C6 arising from the entoconid were only observed in Pongo (10%).

Table 6.

Variation in origin of LM C6 of lower molars

Taxon Total Absent, % Present, n (%) Distal fovea, % Hypoconulid, % Entoconid, % Occlusal surface, % Enamel, %
A. anamensis 8 62.5 3 (37.5) 100.0 0.0 0.0 0.0 0.0
A. afarensis 12 66.7 4 (33.3) 100.0 0.0 0.0 0.0 0.0
A. africanus 48 64.6 17 (35.4) 88.2 0.0 0.0 5.9 5.9
P. aethiopicus 2 100.0 0 (0.0) – – – – –
P. boisei 6 0.0 6 (100.0) 100.0 0.0 0.0 0.0 0.0
P. robustus 50 20.0 40 (80.0) 97.5 0.0 0.0 0.0 2.5
Homo sp./habilis s.l. 10 60.0 4 (40.0) 100.0 0.0 0.0 0.0 0.0
H. erectus s.l. 6 100.0 0 (0.0) – – – – –
MPAE 3 66.7 1 (33.3) 100.0 0.0 0.0 0.0 0.0
H. neanderthalensis 64 54.7 29 (45.3) 75.9 24.1 0.0 0.0 0.0
H. sapiens (Pleistocene) 40 82.5 7 (17.5) 100.0 0.0 0.0 0.0 0.0
H. sapiens (recent) 167 89.8 17 (10.2) 100.0 0.0 0.0 0.0 0.0
P. troglodytes 94 35.1 61 (64.9) 65.6 31.1 0.0 3.3 0.0
P. paniscus 24 75.0 6 (25.0) 100.0 0.0 0.0 0.0 0.0
Gorilla sp. 34 88.2 4 (11.8) 75.0 0.0 0.0 25.0 0.0
Pongo sp. 44 77.3 10 (22.7) 80.0 0.0 10.0 10.0 0.0
Open in a new tab

MPAE, Middle Pleistocene African and European humans.

Frequencies of origin types per taxon do not include individuals with trait absence.

Figure 5.

Figure 5

Open in a new tab

Types of LM C6 development. (A) LM C6 absent [enamel–dentine junction (EDJ) of recent H. sapiens illustrated]; (B) distal fovea type (EDJ of P. t. verus illustrated); (C) hypoconulid type (EDJ of P. t. verus illustrated); (D) entoconid type (EDJ of Pongo illustrated); (E) occlusal type (EDJ of Pongo illustrated); and (F) enamel type [EDJ of A. africanus STW 412a LRM2 illustrated with distal view of outer enamel surface (OES) inset in bottom right corner]. Lingual to the left.

Examination at the EDJ shows that cases of two or more ‘LM C6’ dentine horns are not rare in hominoids, with frequencies ranging from 2.2% to 27.3% (Table S5). For samples with more than 10 observations, absence of this feature was only found in A. afarensis, H. habilis s.l. and H. sapiens (both Pleistocene and recent). Multiple LM C6s usually arise from the same structure and rarely from different structures. Except for one H. neanderthalensis and one P. troglodytes showing a LM ‘double’ C6 entirely arising from the hypoconulid, all cases of multiple ‘LM C6’ dentine horns originating from the same structure were associated with the distal fovea.

Cusp 7 of the lower molars

All taxa show a high and significant correlation in LM C7 expressions at the EDJ and OES (Table 7). Correlation coefficients range from 0.73 to 1, with the lowest values found in recent H. sapiens (r = 0.73), A. africanus (r = 0.75) and P. robustus (r = 0.77). However, discrepancies are not substantial, and in the majority of cases involve the ‘suspected’ category. Major sources of discrepancy include the presence of a marked (and sometimes pointed) shouldering on the distal slope of the metaconid as revealed at the EDJ, which may or may not be associated with a clear LM C7 at the OES. This is particularly the case in H. sapiens (Pleistocene and recent), P. paniscus and P. troglodytes. Except for one A. africanus and one recent H. sapiens specimen, there is no evidence of LM C7 formed entirely by enamel deposition. In these two cases, each molar shows one LM C7 dentine horn but two small cuspules associated with this cusp at the OES.

Table 7.

Spearman's correlation coefficient test for LM C7 expression at the EDJ and OES

Taxon n r P‐value
East African Australopithecus a 20 1.000 < 0.001
A. africanus 38 0.747 < 0.001
P. robustus 31 0.765 < 0.001
Early Homo b 15 0.933 < 0.001
H. neanderthalensis 42 0.963 < 0.001
Pleistocene H. sapiens 37 0.911 < 0.001
Recent H. sapiens 139 0.729 < 0.001
P. troglodytes 94 0.823 < 0.001
P. paniscus 24 0.797 < 0.001
Gorilla sp. 34 0.886 < 0.001
Pongo sp. 44 0.826 < 0.001
Open in a new tab

Significant values in bold.

a

Includes A. anamensis and A. afarensis.

b

Includes H. habilis s.l. and H. erectus s.l.

From the 665 hominoid molars examined, LM C7 is only present in 130 (19.5%) teeth. Frequencies of LM C7 morphological types per taxon following Skinner et al.'s (2008) criteria are given in Table S6. Variation in LM C7 origin is summarized in Table 8 and Fig. 6. This study supports Skinner et al. (2008), who suggested that LM C7 can form from either the distal shoulder of the metaconid (Skinner's metaconulid type) or the lingual groove (Skinner's interconulid type). Three additional variants were identified here such that LM C7 can derive from the mesial slope of the entoconid, the occlusal surface or from enamel deposition alone. However, these variants rarely occur and can be considered exceptions to the most common manifestations proposed by Skinner et al. (2008). The LM C7 most frequently arises from the metaconid in Australopithecus, Paranthropus, Pan and recent H. sapiens, whereas it is most commonly associated with the lingual groove in Homo (except for recent H. sapiens), Gorilla and Pongo. Some additional subtle patterns include the high occurrence of LM C7 in H. habilis s.l. (62.5%) compared with other groups, as well as the large number of molars with shouldering on the metaconid in Pan (P. troglodytes and P. paniscus) and to a lesser extent in H. sapiens (Pleistocene and recent). This contrasts with most other taxa examined, which generally exhibit a smooth surface on the metaconid when LM C7 is absent (Table 8). Noteworthy is that the metaconid shouldering does not necessarily represent an earlier or interrupted stage of LM C7 formation, as shouldering can also occur in conjunction with the clear presence of this cusp. Turner et al.'s (1991) ASUDAS also included an indecisive category (‘grade 1A: a faint tip‐less cusp 7 occurs displaced as a bulge on the lingual surface of cusp 2’ p. 24) for LM C7 at the OES, which appears to correspond to the shouldering type observed at the EDJ. Analyses of the EDJ also show that molars with LM ‘double’ C7s are extremely rare (Table S7). Only one chimpanzee presents this feature among the more than 650 hominoid teeth studied.

Table 8.

Variation in origin of LM C7 of lower molars

Taxon Total Absent, % Present, n (%) Metaconid, % Lingual groove, % Entoconid, % Occlusal surface, % Enamel, %
A. anamensis 9 88.9 1 (11.1) 0.0 0.0 100.0 0.0 0.0
A. afarensis 11 81.8 2 (18.2) 50.0 50.0 0.0 0.0 0.0
A. africanus 60 65.0 21 (35.0) 76.2 23.8 0.0 0.0 0.0
P. aethiopicus 2 100.0 0 (0.0) – – – – –
P. boisei 7 85.7 1 (14.3) 100.0 0.0 0.0 0.0 0.0
P. robustus 62 79.0 13 (21.0) 69.2 23.1 0.0 0.0 7.7
Homo sp./habilis s.l. 8 37.5 5 (62.5) 20.0 80.0 0.0 0.0 0.0
H. erectus s.l. 7 71.4 2 (28.6) 50.0 50.0 0.0 0.0 0.0
MPAE 3 100.0 0 (0.0) – – – – –
H. neanderthalensis 77 71.4 22 (28.6) 36.4 59.1 0.0 4.5 0.0
H. sapiens (Pleistocene) 42 71.4 12 (28.6) 25.0 75.0 0.0 0.0 0.0
H. sapiens (recent) 169 95.9 7 (4.1) 85.7 14.3 0.0 0.0 0.0
P. troglodytes 103 79.6 21 (20.4) 81.0 14.3 4.8 0.0 0.0
P. paniscus 25 68.0 8 (32.0) 75.0 25.0 0.0 0.0 0.0
Gorilla sp. 32 59.4 13 (40.6) 15.4 84.6 0.0 0.0 0.0
Pongo sp. 48 95.8 2 (4.2) 0.0 100.0 0.0 0.0 0.0
Open in a new tab

MPAE, Middle Pleistocene African and European humans.

Frequencies of origin types per taxon do not include individuals with trait absence.

Figure 6.

Figure 6

Open in a new tab

Types of LM C7 development (lingual view). (A) Skinner et al.'s (2008) type A: no C7 manifestation at the enamel–dentine junction (EDJ) (EDJ of P. troglodytes illustrated); (B) Skinner et al.'s (2008) type B: moderately pronounced shoulder on the distal ridge of the metaconid dentine horn (EDJ of P. troglodytes illustrated); (C) Skinner et al.'s (2008) type C: metaconulid‐type C7 on the distal shoulder of the metaconid DH, which in some cases can resemble a small dentine horn‐like feature (EDJ of P. troglodytes illustrated); (D) Skinner et al.'s (2008) type D: metaconulid‐ type C7 in which a dentine horn‐like feature is not closely associated with the metaconid dentine horn (EDJ of recent H. sapiens illustrated); (E) Skinner et al.'s (2008) type E: interconulid‐type C7 with a dentine horn‐like feature on the distal ridge (but separated from the shoulder by a trough) of the metaconid dentine horn; (EDJ of Homo sp. DNH 67 LRM1 illustrated); (F) Skinner et al.'s (2008) type F: interconulid‐type C7 with a dentine horn‐like feature at the low point on the ridge between the metaconid dentine horn and entoconid dentine horn (EDJ of H. neanderthalensis SD 780 LLM1 illustrated mirror‐imaged); (G) occlusal type (EDJ of H. neanderthalensis Vi 11–39 LRM3 illustrated); and entoconid type (EDJ of P. troglodytes illustrated). Mesial to the left.

Discussion

The use of marginal tubercles on the paracone as taxonomic markers has been overlooked, likely because they are difficult to detect at the OES. This study has demonstrated that the PPT is highly distinctive of some hominoid groups. From the 64 Neandertal upper molars examined, only one specimen shows no traces of PPT. More than 70% of the Neandertal sample exhibits a PPT with either a clear or blunt tubercle next to the tip of the cusp. This appears to be derived in H. neanderthalensis (and possibly MPAE based on the five individuals examined) relative to ancestral condition seen in earlier hominins, which show a smooth surface on the distal slope of the paracone. The occasional presence of this trait in Pleistocene and, to a lesser extent, recent H. sapiens, also suggests that this taxon likely inherited the developmental predisposition for PPT from its LCA with H. neanderthalensis. This is supported by a recent 3D cranial reconstruction of the hypothetical LCA of H. neanderthalensis and H. sapiens by Mounier & Lahr (2016), who suggested that this LCA was more similar to H. neanderthalensis. Under this scenario, the presence of PPT in H. sapiens may have been gradually lost through genetic drift and periods of drastic demographic change. The nearly ubiquitous presence of PPT in H. neanderthalensis compared with contemporaneous H. sapiens adds to the taxonomically informative morphological features identified by Bailey (2002, 2006) for differentiating the upper molars of these two taxa.

As expected given the diverse origins for the hypocone and other accessory cusps (Gregory, 1922; Butler, 1952, 1956, 1978; Hunter & Jernvall, 1995; Jernvall, 1995) and expanding on the findings of Skinner et al. (2008, 2014), this study shows that cusp 5 of the upper molars and cusps 6 and 7 of the lower molars can form in a variety of ways, in association with different dental tissues (i.e. oral epithelium, enamel matrix) and dental structures. Within‐ and between‐group variability in trait origin is highest in UM C5, followed by LM C7 and finally LM C6, which shows a clear tendency across all hominoids to arise as an outgrowth of the distal fovea. This high degree of variability poses problems to a landmark‐based approach to study evolutionary novelties. The inconsistency of cusp origin both within and between species unfortunately makes Klingenberg's (2008) proposed method of landmark duplication unsuitable for morphological innovations associated with hominoid molar shape.

There is no universal criterion for identifying homology and homoplasy, and although most evolutionary biologists today define these terms with respect to a phylogenetic tree, as a post hoc definition, this requires a good understanding of the evolutionary relationships among members of a clade and their patterns of character distribution (Simpson, 1961; Hennig, 1966; Hall, 1994; Rieppel, 1994; Lockwood & Fleagle, 1999). This is especially challenging when applied to the fossil record, and in particular to hominin evolution. Furthermore, it has been argued that it is impossible to attribute morphological similarities to common ancestry without a clear understanding of other causes that may lead to these similarities (Lieberman, 1999). In order to overcome these limitations, alternative definitions of homology beyond the phylogenetic framework have been proposed over the past decades. Among them, the concept of developmental homology has been of main interest, which can be broadly defined as the sharing of common developmental processes (Roth, 1984; Lieberman, 1999). The lack of any species‐specific patterns of UM C5, LM C6 and LM C7 origin suggests that accessory cusps in hominoids are developmentally homoplastic and that they may not be useful for identifying phylogenetic homology. It should be noted, however, that these two concepts are not always mutually exclusive, nor must they agree with each other such that phylogeny‐based homologous characters could be the result of different developmental processes and vice versa (Lieberman, 1999; Lockwood & Fleagle, 1999). The fact that dental traits are equally or more homoplastic than cranial and postcranial characters in primates and other mammals, and that trees based on dental and skeletal characters are not always consistent with those derived from molecular data has been documented extensively (Kay, 1990; Harris & Disotell, 1998; Ross et al. 1998; Sánchez‐Villagra & Williams, 1998). A recent study by Riga et al. (2014) in recent humans also found that environmental stress significantly increases the morphological variability and number of cusps within a tooth. Therefore, we highlight the need of using different lines of evidence for character definition and selection, which ultimately will lead to more reliable phylogenetic reconstructions.

Whether the approach followed here can be used as a valid criterion for identifying homoplastic features at different taxonomic levels has been the subject of debate (Simpson, 1955; Butler, 1956, 1963, 2000; Van Valen, 1994), but we suggest caution when using accessory cusps for assessing hominoid evolutionary relationships as cusps that look similar at the external surface may have originated from different dental structures or tissues. This issue is particularly clear for cases of ‘enamel cusps/cuspules’, which are entirely the product of enamel formation and appear later in tooth development compared with those associated with the growth and folding of the inner enamel epithelium. Cases of enamel‐derived accessory cusps/cuspules are rare in hominoids (and other mammals), although they appear in moderate frequencies in Paranthropus and, to a lesser extent, Australopithecus upper molars. Identifying developmental homoplasy among accessory cusps formed prior to ameloblast–odontoblast differentiation appears to be more difficult, however, as cusps that look similar based on topological associations may be the product of different developmental factors that can be altered in various ways to produce the ‘same’ cusp. Computational modelling, experimental testing on model organisms and population‐level variation in seals point to two key factors determining cusp presence, size and position‐formalized under the patterning cascade model: (i) the overall duration of crown morphogenesis (and thus the size of the inner enamel epithelium); and (ii) the spatiotemporal pattern of enamel knot activation and silencing (Jernvall, 1995, 2000; Jernvall & Jung, 2000; Salazar‐Ciudad & Jernvall, 2002, 2010). Enamel knots are non‐proliferative epithelial cells that produce both activator and inhibitor molecules of signaling pathways such that they prevent the formation of additional enamel knots within the inhibition zone of the surrounding area. Only when escaping this inhibition field can a new enamel knot (and thus new cusp) form. Using crown size and cusp spacing as a proxy for the duration of crown morphogenesis and inhibition field size, respectively, studies of cusp 6 in Pan (Skinner & Gunz, 2010) and Carabelli's cusp in recent humans (Hunter et al. 2010; Moormann et al. 2013) have supported the applicability of the model in hominoids and highlighted the critical role of intercusp spacing relative to tooth size. Given that enamel knots appear sequentially at the tip of the future cusps and that the effects of enamel knot formation within the developing tooth are cumulative, evolutionary developmental biologists have already suggested that later‐forming cusps (as the accessory cusps studied here) are not only expected to be more variable, but also more likely to be homoplastic, which is consistent with the results presented here (Jernvall, 1995, 2000; Jernvall & Jung, 2000; but see also Polly, 1998). Not only can accessory cusps develop in a variety of ways, but also their occurrence within the hominoid lineage appears to be highly variable. Although this variability makes the accessory cusps an important source of information for population‐level studies of recent humans as demonstrated by Scott & Turner (1997), it brings into question their utility for assessing evolutionary relationships and discriminating between groups at higher taxonomic levels (Jernvall, 2000; Jernvall & Jung, 2000). Yet, some taxonomically informative patterns are evident: (i) UM C5 in P. troglodytes most frequently arises from the hypocone, which contrasts with the bucco‐central position of the cusp in most hominins; (ii) A. africanus presents a unique pattern where UM C5 most often derives from the buccal cingulum; and (iii) H. habilis s.l. exhibits a notably high frequency of LM C7 at the EDJ (see also Wood & Abbott, 1983 for corresponding frequencies at the OES).

Given their variability and the strong likelihood of homoplasy, this study supports Butler (1978, 1985) and Van Valen (1994) that accessory cusp terminology should only denote topography, without necessarily implying phylogenetic homology. However, the term metaconule, which has frequently been used to refer to UM C5 of the upper molars of recent humans (Harris & Bailit, 1980; Townsend et al. 1986; Turner et al. 1991) may be inappropriate, as noted by Moormann et al. (2013). This term was originally coined to denote a cusp occurring on the crista obliqua (postprotocrista; Szalay, 1969; Rosenberger & Kinzey, 1976) and, in fact, has been argued to be the source of hypocone development in several mammalian taxa (Hunter & Jernvall, 1995). Although the metaconule has been occasionally observed in humans and other hominoids (Hanihara, 1956; Kanazawa et al. 1990), this cusp more frequently occurs in platyrrhines and has also been observed in Oligocene Parapithecus (Rosenberger & Kinzey, 1976). Among the more than 450 upper molars examined, we only found four cases of a true metaconule (one recent H. sapiens, one H. neanderthalensis, one A. africanus and one P. troglodytes). Importantly, the P. troglodytes individual possesses both a metaconule and a small UM C5. The nature of the developmental processes underlying the formation of the metaconule and other occlusal‐derived cusps, as well as whether or not they are similar to those arising from the marginal ridge remains unknown. Furthermore, the presence of UM ‘double’ C5s (which can derive from similar or different structures) renders an additional complicating factor such that it is unclear if one of these cusps should be classified instead as UM C6. And, if so, should it be so named regardless of its developmental origin? Cusp origin at the EDJ also suggests that the use of entoconulid and metaconilud for LM C6 and LM C7, respectively, may not be appropriate either as these terms imply an association of LM C6 with the entoconid and of LM C7 with the metaconid. While LM C7 in most cases does originate from the metaconid, cases of entoconid‐derived LM C6 are infrequent.

Although many questions remain to be answered, the results of this study not only have implications for cusp terminology, but also uncover previously unknown variation in tubercles and accessory cusps of hominoid upper and lower molars. This research also highlights the utility of the EDJ for human evolutionary studies, and demonstrates that features that look similar at the external surface may be the product of different developmental patterns.

Author contributions

AO, SEB, MMS conceived and designed the study. JJH, MMS acquired the data; AO, MMS processed the data; AO analyzed the data. AO wrote the paper with input from all co‐authors.

Supporting information

Table S1 Fossil hominin upper molars used in this study including accession number locality/site and source.

Table S2 Fossil hominin lower molars used in this study including accession number locality/site and source.

Table S3 Frequencies of UM C5 assessed at the EDJ.

Table S4 Results of the bootstrapping analysis (1000 iterations) for PPT expression.

Table S5 Frequencies of LM C6 assessed at the EDJ.

Table S6 Frequencies of LM C7 manifestation at the EDJ following Skinner et al.'s (2008) classification.

Table S7 Frequencies of C7 assessed at the EDJ.

Click here for additional data file. (116.2KB, docx)

Acknowledgements

Access to specimens was kindly provided by the following institutions: Croatian Museum of Natural History, Ditsong National Museum of Natural History, Francisc Rainer Anthropology Institute, National Museums of Kenya, Musée d'Art et d'Archéologie du Périgord, Musée d'Archéologie Nationale de Saint‐Germain‐en‐Laye, Musée d'Angoulême, Musée National de Préhistoire des Eyzies, Musée de l'Homme, Museo Nacional de Ciencias Naturales (Madrid), Max Planck Institute for Evolutionary Anthropology, Royal Museum for Central Africa (Tervuren), Museum für Vor und Frühgeschichte (Berlin), National Museum of Ethiopia, Museum National d'Histoire Naturelle (Paris), American Museum of Natural History (New York), Beaker People Project, Rockefeller Museum, Royal Belgian Institute of Natural Sciences, Senckenberg Research Institute (Frankfurt), Sackler School of Medicine, Tel Aviv University, Leipzig University Anatomical Collection, University of Witwatersrand, Institut National des Sciences du Patrimoine et de l'Archéologie, the Direction du Patrimoine Culturel and the Musée Archéologique de Rabat. The authors also thank Colin Menter for access to material from Drimolen, Bill Kimbel for access to material from Hadar, and Teku Jacob for access to material from Sangiran. The authors thank Heiko Temming, David Plotzki and Lukas Westphal for technical assistance, and Susan Antón, Terry Harrison, Catalina Villamil and Myra Laird for help and comments throughout this project. The authors also thank Anthony Graham and the anonymous reviewers for their useful comments and constructive criticism on previous versions of this manuscript. This research was supported by the National Science Foundation, the Wenner‐Gren Foundation, the Leakey Foundation, the NYU GSAS James Arthur Fellowship, and the Max Planck Society.

References

  1. Anemone RL, Skinner MM, Dirks W (2012) Are there two distinct types of hypocone in Eocene primates? The “pseudohypocone” of northarctines revised. Palaeontol Electronica 15, 26A. [Google Scholar]
  2. Bailey SE (2002) Neandertal dental morphology: implications for modern human origins. Ph.D. Dissertation, Arizona State University.
  3. Bailey SE (2004) A morphometric analysis of maxillary molar crowns of Middle‐Late Pleistocene hominins. J Hum Evol 47, 183–198. [DOI] [PubMed] [Google Scholar]
  4. Bailey SE (2006) Beyond shovel shaped incisors: Neandertal dental morphology in a comparative context. Period Biol 108, 253–267. [Google Scholar]
  5. Bailey SE, Wood BA (2007) Trends in postcanine occlusal morphology within the hominin clade: the case of Paranthropus In: Dental Perspectives on Human Evolution. (eds Bailey SE, Hublin J‐J.), pp. 53–64. Berlin: Springer. [Google Scholar]
  6. Bailey SE, Weaver TD, Hublin J‐J (2009) Who made the Aurignacian and other early Upper Paleolithic industries? J Hum Evol 57, 11–26. [DOI] [PubMed] [Google Scholar]
  7. Butler PM (1939) Studies of the mammalian dentition. Differentiation of the post‐canine dentition. Proc Zool Soc Lond B Syst Morphol 109, 1–36. [Google Scholar]
  8. Butler PM (1952) Molarization of the premolars in the Perissodactyla. Proc Zool Soc Lond 121, 819–843. [Google Scholar]
  9. Butler PM (1956) The ontogeny of molar pattern. Biol Rev Camb Philos Soc 31, 30–69. [Google Scholar]
  10. Butler PM (1963) Tooth morphology and primate evolution In: Dental Anthropology. (ed. Brothwell DR.), pp. 1–13. New York, NY: Pergamon Press. [Google Scholar]
  11. Butler PM (1978) Molar cusp nomenclature and homology In: Development Function and Evolution of Teeth. (eds Butler PM, Joysey KA.), pp. 439–453. London, UK: Academic Press. [Google Scholar]
  12. Butler PM (1985) Homologies of molar cusps and crests and their bearing on assessments of rodent phylogeny In: Evolutionary Relationships Among Rodents. (eds Luckett WP, Hartenberger JL.), pp. 381–401. New York, NY: Plenum Press. [Google Scholar]
  13. Butler PM (1990) Early trends in the evolution of tribosphenic molars. Biol Rev 65, 529–552. [Google Scholar]
  14. Butler PM (2000) The evolution of tooth shape and tooth function in primates In: Development Function and Evolution of Teeth. (eds Teaford M, Smith M, Ferguson M.), pp. 201–211. Cambridge: Cambridge University Press. [Google Scholar]
  15. Corruccini RS (1987) The dentinoenamel junction in primates. Int J Primatol 8, 99–114. [Google Scholar]
  16. Corruccini RS, Holt BM (1989) The dentinoenamel junction and the hypocone in primates. Hum Evol 4, 253–262. [Google Scholar]
  17. Gregory W (1922) The Origin and Evolution of the Human Dentition. Baltimore, MD: Williams and Wilkins. [Google Scholar]
  18. Guatelli‐Steinberg D, Irish JD (2005) Brief communication: early hominin variability in first molar dental trait frequencies. Am J Phys Anthropol 128, 477–484. [DOI] [PubMed] [Google Scholar]
  19. Hall BK. (ed.) (1994) Homology: the Hierarchical Basis of Comparative Biology. San Diego, CA: Academic Press. [Google Scholar]
  20. Hammer Ø, Harper DAT, Ryan PD (2001) PAST: paleontological statistics software package for education and data analysis. Palaeontol Electronica 4, 1–9. [Google Scholar]
  21. Hanihara K (1956) Studies on the deciduous dentition of the Japanese and the Japanese‐American hybrids IV. Deciduous upper molars. J Anthropol Soc Nippon 65, 67–87. [Google Scholar]
  22. Harris EF, Bailit HL (1980) The metaconule: a morphologic and familial analysis of a molar cusp in humans. Am J Phys Anthropol 53, 349–358. [DOI] [PubMed] [Google Scholar]
  23. Harris EE, Disotell TR (1998) Nuclear gene trees and the phylogenetic relationships of the mangabeys (Primates: Papionini). Mol Biol Evol 15, 892–900. [DOI] [PubMed] [Google Scholar]
  24. Hennig W (1966) Phylogenetic Systematics. Urbana, IL: University of Illinois Press. [Google Scholar]
  25. Hershkovitz P (1971) Basic crown patterns and cusp homologies of mammalian teeth In: Dental Morphology and Evolution. (ed. Dahlberg A.), pp. 95–150. Chicago, IL: Chicago University Press. [Google Scholar]
  26. Hershkovitz P (1977) Living New World Monkeys Platyrrhini. With an Introduction to Primates, Vol. 1. Chicago, IL: University of Chicago Press. [Google Scholar]
  27. Hunter JP, Jernvall J (1995) The hypocone as a key innovation in mammalian evolution. Proc Natl Acad Sci USA 92, 10 718–10 722. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Hunter JP, Guatelli‐Steinberg D, Weston TC, et al. (2010) Model of tooth morphogenesis predicts Carabelli cusp expression, size, and symmetry in humans. PLoS One 5, 1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Irish JD, Guatelli‐Steinberg D, Legge SS, et al. (2013) Dental morphology and the phylogenetic “place” of Australopithecus sediba . Science 340, 1 233 062–1 233 064. [DOI] [PubMed] [Google Scholar]
  30. Jernvall J (1995) Mammalian molar cusp patterns: developmental mechanisms of diversity. Acta Zool Fenn 198, 1–61. [Google Scholar]
  31. Jernvall J (2000) Linking development with generation of novelty in mammalian teeth. Proc Natl Acad Sci USA 97, 2641–2645. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Jernvall J, Jung H‐S (2000) Genotype phenotype and developmental biology of molar tooth characters. Yearb Phys Anthropol 43, 171–190. [DOI] [PubMed] [Google Scholar]
  33. Jernvall J, Gilbert CC, Wright PC (2008) Peculiar tooth homologies of the greater bamboo lemur Prolemur = Hapalemur simus. When is a Paracone not a Paracone? In: Elwyn Simons: A Search for Origins. (eds Fleagle JG, Gilbert CC.), pp. 335–342. New York, NY: Springer Science and Business Media. [Google Scholar]
  34. Kanazawa E, Seikikawa M, Ozaki T (1990) A quantitative investigation of irregular cuspules in human maxillary permanent molars. Am J Phys Anthropol 83, 173–180. [DOI] [PubMed] [Google Scholar]
  35. Kay RF (1990) The phyletic relationships of extant and fossil Pitheciinae (Platyrrhini, Anthropoidea). J Hum Evol 19, 175–208. [Google Scholar]
  36. Klingenberg CP (2008) Novelty and “homology‐free” morphometrics: what's in a name? Evol Biol 35, 186–190. [Google Scholar]
  37. Korenhof CAW (1960) Morphogenetical Aspects of the Human Upper Molar. Utrecht: Uitgeversmaatschappij Neerlandia. [Google Scholar]
  38. Kraus BS (1952) Morphologic relationships between enamel and dentin surfaces of lower first molar teeth. J Dent Res 31, 248–256. [DOI] [PubMed] [Google Scholar]
  39. Lieberman DE (1999) Homology and hominid phylogeny: problems and potential solutions. Evol Anthropol 7, 142–151. [Google Scholar]
  40. Lockwood CA, Fleagle JG (1999) The recognition and evaluation of homoplasy in primate and human evolution. Yearb Phys Anthropol 42, 189–232. [DOI] [PubMed] [Google Scholar]
  41. Luo Z‐X, Cifelli RL, Kielan‐Jaworowska Z (2001) Dual origins of tribosphenic mammals. Nature 409, 53–57. [DOI] [PubMed] [Google Scholar]
  42. Martin RMG, Hublin J‐J, Gunz P, et al. (2017) The morphology of the enamel‐dentine junction in Neanderthal molars: gross morphology non‐metric traits and temporal trends. J Hum Evol 103, 20–44. [DOI] [PubMed] [Google Scholar]
  43. Martinón‐Torres M, Bermúdez de Castro JM, Gómez‐Robles A, et al. (2007) Dental evidence on the hominin dispersals during the Pleistocene. Proc Natl Acad Sci USA 104, 13 279–13 282. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Martinón‐Torres M, Bermúdez de Castro JM, Gómez‐Robles A, et al. (2008) Dental remains from Dmanisi Republic of Georgia: morphological analysis and comparative study. J Hum Evol 55, 249–273. [DOI] [PubMed] [Google Scholar]
  45. Martinón‐Torres M, Bermúdez de Castro JM, Gómez‐Robles A, et al. (2012) Morphological description and comparison of the dental remains from Atapuerca‐Sima de los Huesos site (Spain). J Hum Evol 62, 7–58. [DOI] [PubMed] [Google Scholar]
  46. Molnar S (1971) Human tooth wear tooth function and cultural variability. Am J Phys Anthropol 34, 27–42. [DOI] [PubMed] [Google Scholar]
  47. Moormann S, Guatelli‐Steinberg D, Hunter J (2013) Metamerism, morphogenesis, and the expression of Carabelli and other dental traits in humans. Am J Phys Anthropol 150, 400–408. [DOI] [PubMed] [Google Scholar]
  48. Morita W, Morimoto N, Ohshima H (2016) Exploring metameric variation in human molars: a morphological study using morphometric mapping. J Anat 229, 343–355. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Mounier A, Lahr MM (2016) Virtual ancestor reconstruction: revealing the ancestor of modern humans and Neandertals. J Hum Evol 91, 57–92. [DOI] [PubMed] [Google Scholar]
  50. Ortiz A, Skinner MM, Bailey SE, et al. (2012) Carabelli's trait revisited: an examination of mesiolingual features at the enamel‐dentine junction and enamel surface of Pan and Homo sapiens upper molars. J Hum Evol 63, 586–596. [DOI] [PubMed] [Google Scholar]
  51. Patterson B (1956) Early Cretaceous mammals and the evolution of mammalian molar teeth. Fieldiana 13, 1–105. [Google Scholar]
  52. Patterson C (1982) Morphological characters and homology In: Problems of Phylogenetic Reconstruction. (eds Joysey KA, Friday AE.), pp. 21–74. London, UK: Academic Press. [Google Scholar]
  53. Pilbrow VC (2003) Dental variation in African apes with implications for understanding patterns of variation in fossil apes. Ph.D. Dissertation New York University.
  54. Polly PD (1998) Variability, selection, and constraints: development and evolution in viverravid, Carnivora, Mammalia molar morphology. Paleobiology 24, 409–429. [Google Scholar]
  55. R Core Team (2012) R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; http://www.R-project.org/. [Google Scholar]
  56. Rieppel O (1994) Homology, topology and typology: the history of modern debates In: Homology. The Hierarchical Basis of Comparative Biology. (ed. Hall BK.), pp. 64–101. San Diego, CA: Academic Press. [Google Scholar]
  57. Riga A, Belcastro MG, Moggi‐Cecchi J (2014) Environmental stress increases variability in the expression of dental cusps. Am J Phys Anthropol 153, 397–407. [DOI] [PubMed] [Google Scholar]
  58. Rosenberger AL, Kinzey WG (1976) Functional patterns of molar occlusion in platyrrhine primates. Am J Phys Anthropol 45, 281–298. [DOI] [PubMed] [Google Scholar]
  59. Ross C, Williams B, Kay RF (1998) Phylogenetic analysis of anthropoid relationships. J Hum Evol 35, 221–306. [DOI] [PubMed] [Google Scholar]
  60. Roth VL (1984) On homology. Biol J Linnean Soc 22, 13–29. [Google Scholar]
  61. Salazar‐Ciudad I, Jernvall J (2002) A gene network model accounting for development and evolution of mammalian teeth. Proc Natl Acad Sci USA 99, 8116–8120. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Salazar‐Ciudad I, Jernvall J (2010) A computational model of teeth and the developmental origins of morphological variation. Nature 464, 583–588. [DOI] [PubMed] [Google Scholar]
  63. Sánchez‐Villagra MR, Kay RF (1996) Do Phalangeriforms Marsupialia: Diprotodontia have a “hypocone”? Aust J Zool 44, 461–467. [Google Scholar]
  64. Sánchez‐Villagra MR, Williams BA (1998) Levels of homoplasy in the evolution of the mammalian skeleton. J Mamm Evol 5, 113–126. [Google Scholar]
  65. Schour I, Massler M (1940) Studies in tooth development. The growth pattern of the human teeth. J Am Dent Assoc 27, 1778–1793, 1918–1931. [Google Scholar]
  66. Scott GR, Turner CG (1997) The Anthropology of Modern Human Teeth: Dental Morphology and its Variation in Recent Human Populations. Cambridge: Cambridge University Press. [Google Scholar]
  67. Simpson GG (1955) The Phenacolemuridae new family of early Primates. Bull Am Mus Nat Hist 105, 411–442. [Google Scholar]
  68. Simpson GG (1961) Principles of Animal Taxonomy. New York, NY: Columbia University Press. [DOI] [PubMed] [Google Scholar]
  69. Skinner MM (2008) Enamel‐dentine junction morphology of extant hominoid and fossil hominin lower molars. Ph.D. Dissertation, George Washington University.
  70. Skinner MM, Gunz P (2010) The presence of accessory cusps in chimpanzee lower molars is consistent with a patterning cascade model of development. J Anat 217, 245–253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Skinner MM, Wood BA, Boesch C, et al. (2008) Dental trait expression at the enamel‐dentine junction of lower molars in extant and fossil hominoids. J Hum Evol 54, 173–186. [DOI] [PubMed] [Google Scholar]
  72. Skinner MM, Ludeman EM, Bailey SE, et al. (2014) Cusp 6 variation in the hominin clade: insights and implications revealed at the enamel‐dentine junction. Am J Phys Anthropol S58, 241–242. [Google Scholar]
  73. Suwa G, White TD, Howell FC (1996) Mandibular postcanine dentition from the Shungura formation Ethiopia: crown morphology taxonomic allocations and Plio‐Pleistocene hominid evolution. Am J Phys Anthropol 101, 247–282. [DOI] [PubMed] [Google Scholar]
  74. Swindler DR (2002) Primate Dentition. New York, NY: Cambridge University Press. [Google Scholar]
  75. Szalay FS (1969) Mixodectidae microsyopidae and the insectivore‐primate transition. Bull Am Mus Nat Hist 140, 193–330. [Google Scholar]
  76. Townsend G, Yamada H, Smith P (1986) The metaconule in Australian Aboriginals: an accessory tubercle on maxillary molar teeth. Hum Biol 58, 851–862. [PubMed] [Google Scholar]
  77. Turner CG, Nichol CR, Scott GR (1991) Scoring procedures for key morphological traits of the permanent dentition: the Arizona State University Dental Anthropology System In: Advances in Dental Anthropology. (eds Kelley MA, Larsen CS.), pp. 13–31. London, UK: Wiley‐Liss. [Google Scholar]
  78. Uchida A (1996) Craniodental Variation Among the Great Apes. Peabody Museum of Archaeology and Ethnology. Cambridge: Harvard University. [Google Scholar]
  79. Van Valen LM (1982) Homology and causes. J Morphol 173, 305–312. [DOI] [PubMed] [Google Scholar]
  80. Van Valen LM (1994) Serial homology: the crests and cusps of mammalian teeth. Acta Palaeontol Pol 38, 145–158. [Google Scholar]
  81. Wollny G, Kellman P, Ledesma‐Carbayo M‐J, et al. (2013) MIA – a free and open source software for gray scale medical image analysis. Source Code Biol Med 8, 20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  82. Wood BA, Abbott SA (1983) Analysis of the dental morphology of Plio‐Pleistocene hominids. I. Mandibular molars: crown area measurements and morphological traits. J Anat 136, 197–219. [PMC free article] [PubMed] [Google Scholar]
  83. Wood BA, Engleman CA (1988) Analysis of the dental morphology of Plio‐Pleistocene hominids V. Maxillary postcanine tooth morphology. J Anat 161, 1–35. [PMC free article] [PubMed] [Google Scholar]
  84. Wood BA, Abbott SA, Graham SH (1983) Analysis of the dental morphology of Plio‐Pleistocene hominids II. Mandibular molars: study of cusp areas fissure pattern and cross sectional shape of the crown. J Anat 137, 287–314. [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Table S1 Fossil hominin upper molars used in this study including accession number locality/site and source.

Table S2 Fossil hominin lower molars used in this study including accession number locality/site and source.

Table S3 Frequencies of UM C5 assessed at the EDJ.

Table S4 Results of the bootstrapping analysis (1000 iterations) for PPT expression.

Table S5 Frequencies of LM C6 assessed at the EDJ.

Table S6 Frequencies of LM C7 manifestation at the EDJ following Skinner et al.'s (2008) classification.

Table S7 Frequencies of C7 assessed at the EDJ.

Click here for additional data file. (116.2KB, docx)

Articles from Journal of Anatomy are provided here courtesy of Anatomical Society of Great Britain and Ireland

RESOURCES