Evolutionary Relationships Among Robust and Gracile Australopiths: An “Evo-devo” Perspective

Hominin fossils of gracile and robust australopith groups were found both in East and in South Africa. It is unclear, however, whether all robusts belong to a monophyletic Paranthropus clade, as the craniofacial resemblance among robust australopiths might only be a superficial correlate of similar masticatory adaptations and not evidence of shared ancestry. It has been suggested that the East African Australopithecus/Paranthropus boisei and the South African A./P. robustus might be convergent allometric variants of their gracile geographical neighbors A. afarensis and A. africanus. Here we approach the phylogenetic questions about robust and gracile australopiths from an ??evo-devo?? perspective, examining how simple alterations of development could contribute to the shape differences among hominin species. Using geometric morphometrics we compare gracile and robust australopith crania in the context of the allometric scaling patterns of Pan troglodytes, P. paniscus, and Gorilla gorilla. We examine support for two alternative evolutionary scenarios based on predictions derived from quantitative genetics models: either (1) A./P. robustus evolved in South Africa from the gracile A. africanus, or (2) A./P. robustus is a local variant of the eastern African A./P. boisei. We use developmental simulations to demonstrate that some robust characteristics (wide faces, anteriorly placed zygomatics, and facial dishing) can be predicted by allometric scaling along the ontogenetic trajectory of the gracile A. africanus. We find, however, that the facial differences between A. africanus specimens (Taung, Sts 5, Sts 71, and Stw 505) and A./P. robustus specimen SK 48 cannot be explained by allometric scaling alone. Facial shape differences between A./P. robustus SK 48 and A./P. boisei (KNM-ER 732, KNM-ER 406, OH 5) and the A./P. aethiopicus specimen KNM-WT 17000, on the other hand, can largely be explained by allometric scaling. This is consistent with a close evolutionary relationship of these robust taxa.     

Dental Ontogeny in Pliocene and Early Pleistocene Hominins

Until recently, our understanding of the evolution of human growth and development derived from studies of fossil juveniles that employed extant populations for both age determination and comparison. This circular approach has led to considerable debate about the human-like and ape-like affinities of fossil hominins. Teeth are invaluable for understanding maturation as age at death can be directly assessed from dental microstructure, and dental development has been shown to correlate with life history across primates broadly. We employ non-destructive synchrotron imaging to characterize incremental development, molar emergence, and age at death in more than 20 Australopithecus anamensis, Australopithecus africanus, Paranthropus robustus and South African early Homo juveniles. Long-period line periodicities range from at least 6–12 days (possibly 5–13 days), and do not support the hypothesis that australopiths have lower mean values than extant or fossil Homo. Crown formation times of australopith and early Homo postcanine teeth fall below or at the low end of extant human values; Paranthropus robustus dentitions have the shortest formation times. Pliocene and early Pleistocene hominins show remarkable variation, and previous reports of age at death that employ a narrow range of estimated long-period line periodicities, cuspal enamel thicknesses, or initiation ages are likely to be in error. New chronological ages for SK 62 and StW 151 are several months younger than previous histological estimates, while Sts 24 is more than one year older. Extant human standards overestimate age at death in hominins predating Homo sapiens, and should not be applied to other fossil taxa. We urge caution when inferring life history as aspects of dental development in Pliocene and early Pleistocene fossils are distinct from modern humans and African apes, and recent work has challenged the predictive power of primate-wide associations between hominoid first molar emergence and certain life history variables.     

Endocranial volume of Australopithecus africanus: new CT-based estimates and the effects of missing data and small sample size

Estimation of endocranial volume in Australopithecus africanus is important in interpreting early hominin brain evolution. However, the number of individuals available for investigation is limited and most of these fossils are, to some degree, incomplete and/or distorted. Uncertainties of the required reconstruction (‘missing data uncertainty’) and the small sample size (‘small sample uncertainty’) both potentially bias estimates of the average and within-group variation of endocranial volume in A. africanus.We used CT scans, electronic preparation (segmentation), mirror-imaging and semilandmark-based geometric morphometrics to generate and reconstruct complete endocasts for Sts 5, Sts 60, Sts 71, StW 505, MLD 37/38, and Taung, and measured their endocranial volumes (EV). To get a sense of the reliability of these new EV estimates, we then used simulations based on samples of chimpanzees and humans to: (a) test the accuracy of our approach, (b) assess missing data uncertainty, and (c) appraise small sample uncertainty.Incorporating missing data uncertainty of the five adult individuals, A. africanus was found to have an average adult endocranial volume of 454-461 ml with a standard deviation of 66-75 ml. EV estimates for the juvenile Taung individual range from 402 to 407 ml. Our simulations show that missing data uncertainty is small given the missing portions of the investigated fossils, but that small sample sizes are problematic for estimating species average EV. It is important to take these uncertainties into account when different fossil groups are being compared.     

The mid-face of lower Pleistocene hominins and its bearing on the attribution of SK 847 and StW 53

SK 847 and StW 53 have often been cited as evidence for early Homo in South Africa. To examine whether midfacial morphology is in agreement with these attributions, we analyze Euclidean distances calculated from 3-D coordinates on the maxillae of SK 847 and StW 53, as well as Australopithecus africanus (Sts 5, Sts 71), Paranthropus robustus (SK 46, SK 48, SK 52, SK 83), early Homo (KNM-ER 1813, KNM-ER 1805, KNM-ER 3733, KNM-WT 15000), P. boisei (KNM-ER 406, KNM-WT 17000, KNM-WT 17400), Gorilla gorilla (n = 116), Homo sapiens (n = 342), Pan paniscus (n = 21) and P. troglodytes (n = 65). Multivariate analyses separate extant hominoids suggesting we have captured taxonomic affinity. With the exception of SK 847 and SK 52, South African fossils tend to cluster together. P. robustus differs substantially from East African robust megadonts. SK 847 and StW 53 resemble the East African Homo specimens that are the most australopith-like, such as KNM-WT 15000 and KNM-ER 1813. The resemblance between StW 53 and Homo is driven partly by similarities in maxillary size. When distances are scaled, StW 53 aligns with A. africanus, while SK 847 clusters primarily with early Homo.     

Upper Molar Cusp Dimensions of South African Australopithecines

Recently, a number of articles have been published on the mechanisms of dental development, and in particular on the genetic control of dental formation. As a specific example of the potential of this kind of research for paleoanthropological studies, cusp areas of upper molars of two species of South African Australopithecines (Australopithecus africanus, A. robustus) have been examined. The results suggest notable differences between the two species, possibly related to markedly different developmental pathways that are still far from clear in their details.     

Individual tooth macrowear pattern guides the reconstruction of Sts 52 (Australopithecus africanus) dental arches

The functional restoration of the occlusal relationship between maxillary and mandibular tooth rows is a major challenge in modern dentistry and maxillofacial surgery. Similar technical challenges are present in paleoanthropology when considering fragmented and deformed mandibular and maxillary fossils. Sts 52, an Australopithecus africanus specimen from Sterkfontein Member 4, represents a typical case where the original shape of the dental arches is no longer preserved. It includes a partial lower face (Sts 52a) and a fragmented mandible (Sts 52b), both incomplete and damaged to such an extent to thwart attempts at matching upper and lower dentitions. We show how the preserved macrowear pattern of the tooth crowns can be used to functionally reconstruct Sts 52's dental arches. High‐resolutiondental stone casts of Sts 52 maxillary and mandibular dentition were mounted and repositioned in a dental articulator. The occlusal relationship between antagonists was restored based on the analysis of the occlusal wear pattern of each preserved tooth, considering all dental contact movements represented in the occlusal compass. The reconstructed dental arches were three‐dimensional surface scanned and their occlusal kinematics tested in a simulation. The outcome of this contribution is the first functional restoration of A. africanus dental arches providing new morphometric data for specimen Sts 52. It is noteworthy that the method described in this case study might be applied to several other fossilspecimens. Am J Phys Anthropol, 2013. © 2013 Wiley Periodicals, Inc.     

Underestimating intraspecific variation: The problem with excluding Sts 19 from Australopithecus africanus

Two analyses conclude that Sts 19 cannot be accommodated within the Australopithecus africanus hypodigm (Kimbel and Rak [1993] In Kimbel and Martin [eds.]: Species, Species Concepts, and Primate Evolution. New York: Plenum, pp. 461–484; Sarmiento [1993] Am. J. Phys. Anthropol. [Suppl.] 16:173). Both studies exclude Sts 19 because it possesses synapomorphies with Homo. Furthermore, according to Kimbel and Rak (1993), including Sts 19 in A. africanus results in an unacceptably high degree of polymorphism. This study aims to refute the null hypothesis that Sts 19 belongs to A. africanus. Twelve basicranial characters, as defined and implemented in Kimbel and Rak's study, were scored for casts of seven A. africanus and seven Homo habilis basicranial specimens. These characters were also examined on specimens from a large (N = 87) sample of African pongids. Contrary to Kimbel and Rak's (1993) findings, the null hypothesis is not refuted. The degree of polymorphism among A. africanus with Sts 19 included is less than that seen in Pan troglodytes. In addition, Sts 19 shares only one apomorphy with Homo. However, when treated metrically, Sts 19's morphology for this character is not significantly divergent from other A. africanus specimens. Am J Phys Anthropol 105:461–480, 1998. © 1998 Wiley-Liss, Inc.     

Continuous dental eruption identifies Sts 5 as the developmentally oldest fossil hominin and informs the taxonomy of Australopithecus africanus

The relatively small Australopithecus africanus specimen Sts 5 has figured prominently in taxonomic debates, and the determination of this specimen as a young male or an elderly female has the potential to offer a great deal of resolution on this question. Sts 5 has been argued to be either a small, immature male or a mature female based on a variety of characters.     

Cross-sectional geometric properties of the Otavipithecus mandible

Cross-sectional geometric properties of the postcanine mandibular corpus are determined for the only known specimen of Otavipithecus namibiensis, a middle Miocene hominoid from southern Africa. It is shown that Otavipithecus is unique in that several important mechanical properties of its mandible, including maximum and minimum moments of inertia and distribution of cortical bone, differ from patterns seen in both extant hominoids and the early hominids Australopithecus africanus and Australopithecus (Paranthropus) robustus. This is particularly apparent in the mechanical design of the posterior portion of the mandibular corpus for resisting increased torsional and transverse bending moments. Cortical index values at the level of M₂ also reveal that both Otavipithecus and A. africanus are similarly designed to resist increased masticatory loads with relatively less cortical bone area, a highly efficient mechanical design. © 1996 Wiley-Liss, Inc.     

Chimpanzee variation facilitates the interpretation of the incisive suture closure in South African Plio-Pleistocene hominids

For a better understanding of early hominid growth patterns, we need to compare skeletal maturation among humans and chimpanzees. This study provides new data on variation of the incisive suture closure in extant species to facilitate the understanding of growth patterns among South African Plio-Pleistocene hominids. The complete anterior closure of the incisive suture occurs early during human life, mostly before birth. In contrast, in chimpanzees a complete anterior closure occurs mostly after the eruption of either the first permanent molars (pygmy chimpanzees) or the third molars (common chimpanzees). The first aim of this study is to test whether the patterns of closure of both the anterior and palatal components of the incisive suture in chimpanzees accurately mirror their polytypism by investigating 720 museum specimens of known geographical origin. Then we use the data gleaned from the incisive suture closure in chimpanzees to determine whether there are different growth patterns among South African Plio-Pleistocene hominids and to interpret them. Results about the pattern of incisive suture closure are consistent with the differences among chimpanzees as revealed by molecular data. Thus, the variation in chimpanzee patterns of incisive suture closure facilitates the interpretation of morphology in South African fossil hominids. In Australopithecus (Paranthropus) robustus as compared to Australopithecus africanus, the complete anterior closure and, probably, the complete palatal closure of the incisive suture occurs during early life in the same way as they occur in humans. Moreover, the closure pattern observed on Stw 53, a supposed early Homo from Sterkfontein Member 5, is similar to that seen in A. africanus and in chimpanzees. Thus, with respect to the anterior component of the incisive suture, A. africanus and Stw 53 retain the primitive feature for which A. (P.) robustus and Homo share the derived character state. Finally, it is worth noting that the Taung child does not show the robust condition. Am J Phys Anthropol 105:121–135, 1998. © 1998 Wiley-Liss, Inc.     

Relative cheek-tooth size in Australopithecus

Until the discovery of Australopithecus afarensis, cheek-tooth megadontia was unequivocally one of the defining characteristics of the australopithecine grade in human evolution along with bipedalism and small brains. This species, however, has an average postcanine area of 757 mm², which is more like Homo habilis (759 mm²) than A. africanus (856 mm²). But what is its relative cheek-tooth size in comparison to body size? One approach to this question is to compare postcanine tooth area to estimated body weight. By this method all Australopithecus species are megadont: they have cheek teeth 1.7 to 2.3 times larger than modern hominoids of similar body size. The series from A. afarensis to A. africanus to A. robustus to A. boisei shows strong positive allometry indicating increasing megadontia through time. The series from H. habilis to H. erectus to H. sapiens shows strong negative allometry which implies a sharp reduction in the relative size of the posterior teeth. Postcanine megadontia in Australopithecus species can also be demonstrated by comparing tooth size and body size in associated skeletons: A. afarensis (represented by A.L. 288–1) has a cheek-tooth size 2.8 times larger than expected from modern hominoids; A. africanus (Sts 7) and A. robustus (TM 1517) are over twice the expected size. The evolutionary transition from the megadont condition of Australopithecus to the trend of decreasing megadontia seen in the Homo lineage may have occurred between 3.0 and 2.5 m.y. from A. afarensis to H. habilis but other evidence indicates that it is more likely to have occurred between 2.5 to 2.0 m.y. from an A. africanus-like form to H. habilis.     

Protostylid expression at the enamel-dentine junction and enamel surface of mandibular molars of Paranthropus robustus and Australopithecus africanus

Distinctive expressions and incidences of discrete dental traits at the outer enamel surface (OES) contribute to the diagnoses of many early hominin taxa. Examination of the enamel-dentine junction (EDJ), imaged non-destructively using micro-computed tomography, has elucidated the morphological development of dental traits and improved interpretations of their variability within and among taxa. The OES expressions of one of these dental traits, the protostylid, have been found to differ among African Plio-Pleistocene fossil hominin taxa. In this study protostylid expression is examined at the OES and at the EDJ of Paranthropus robustus (n = 23) and Australopithecus africanus (n = 28) mandibular molars, with the goals of incorporating EDJ morphology into the definition of the protostylid and assessing the relative contribution of the EDJ and enamel cap to its expression in these taxa. The results provide evidence a) of statistically significant taxon-specific patterns of protostylid morphology at the EDJ that are not evident at the OES; b) that in P. robustus, thick enamel reduces the morphological correspondence between the form of the protostylid seen at the EDJ and the OES, and c) that if EDJ images can be obtained, then the protostylid retains its taxonomic value even in worn teeth.     

Further analysis of mandibular molar crown and cusp areas in Pliocene and early Pleistocene hominids

Crown and cusp areas of mandibular molars were measured and analyzed on a sample of 249 specimens attributed to Australopithecus afarensis, A. africanus, A. (Paranthropus) robustus, A. (P.) boisei, and early Homo. In addition to intertaxon comparisons, we compared data that had been collected independently by two of the authors using methods that differ slightly in technique of measurement. Interobserver differences were evaluated by the t-test of paired comparisons, method error statistic, percent differences, and principal component analysis. Results suggest that between-technique error of measurement of overall crown area is small. Error estimates for individual cusp area measurements were of larger relative magnitude. However, these were not sufficient to detract from the conclusions derived from comparative analyses. Our results are in general agreement with previous assessments of early hominid dental size. Crown areas of A. africanus, however, exhibit a mosaic pattern, with M1 similar in size to that of A. afarensis and early Homo, and M2 and M3 similar in size to that of A. robustus. Intertaxon comparisons of relative cusp area were undertaken by univariate statistics and principal component analysis. These analyses revealed that while A. (P.) robustus and A. (P.) boisei both possess mandibular molars with cusp proportions significantly different from the ‘non-robust’ taxa, these differences are substantially greater in A. (P.) boisei. © 1994 Wiley-Liss, Inc.     

Sex at Sterkfontein: 'Mrs. Ples' is still an adult female

The important question of whether the Australopithecus africanus hypodigm is taxonomically heterogeneous revolves largely around the interpretation of the morphological variation exhibited by the fossils from Sterkfontein. The sex assignment of these specimens is a critical component in the evaluation of their morphological variability. The Sts 5 cranium is pivotal in this regard because it is the most complete and undistorted specimen attributed to A. africanus. Although it has traditionally been regarded as an adult female, this view has been challenged. In particular, it has been argued recently that Sts 5 is a juvenile and that this, together with alveolar bone loss that has supposedly reduced the size of the canine socket, has led to its misinterpretation as a female. Virtual reconstruction of the M³ roots (and/or alveoli) contradicts arguments that these teeth were erupting at the time of death. Regardless, canine emergence and root completion are well ahead of M³ development in juvenile australopiths from Sterkfontein. Thus, even if the M³ root of Sts 5 was incomplete, its canine root would have been fully formed. Measurements of palate depth indicate that the alveolar margins of Sts 5 have not suffered from much (if any) bone loss in the region of the C/P³; any additional bone would result in a palate of truly exceptional depth. Therefore, the dimensions of the canine alveolus of Sts 5 can be regarded as proxies for those of the canine root. The canine root of Sts 5 is among the smallest recorded for any Sterkfontein australopith, which provides strong support for Robert Broom's initial attribution of sex to this specimen. There is no evidence to contradict the assertion that ‘Mrs. Ples’ is an adult female.     

Reappraisal of the taxonomic status of the cranium Stw 53 from the Plio/Pleistocene of Sterkfontein,in South Africa

A fossil skull, Stw 53, from the Plio/Pleistocene of Sterkfontein, in South Africa, has been referred toHomo habilis Leakey, Napier, andTobias, 1964. Reappraisal of its putative hominine affinity reveals a closer resemblance toAustralopithecus africanus Dart, 1925. The skull, as reconstructed, is too small forH. habilis; with no indication of brain expansion overA. africanus; has a facial angle outside the hominine range, but identical with that ofA. africanus; and whose teeth are not elongated but display buccolingual expansion. Although it was found in the same strata (Member 5) as stone tools, there is no causal connection. It has been dated faunistically at 2–1.5 my BP, but due to an unconformity it is suggested that it could be older. In spite of its late date, Stw 53 shows no intermediate characters which could support a trend towardsH. habilis orA. robustus Broom, 1938. It may, therefore, represent a relict population ofA. africanus.     

Several cranial indices and their relevance to fossil man

An evaluation of three cranial indices proposed by Le Gros Clark ('50) is presented in light of hominid fossil material not previously considered. The data analysis suggests that the supra-orbital height index is an important contribution to cranial analysis and elaborations of it may prove valuable. From the results, the authors suggest that the two australopithecine types be separated on the genus level, i.e., Australopithecus africanus and Paranthropus robustus. It is also proposed that Skhūl V be considered Homo sapiens sapiens. The authors generally conclude that a more dynamic approach in primate cranial analysis is needed. This approach should incorporate non-morphological variables.     

Examining affinities of the Taung child by developmental simulation

As a well-preserved juvenile and the type specimen of Australopithecus africanus, the Taung child figures prominently in taxonomic, ontogenetic, and phylogenetic analyses of fossil hominins. Despite general agreement about allocation of Sterkfontein and Makapansgat fossils to this species, limited morphological comparisons have been possible between these adult specimens and the juvenile Taung. Here, we used developmental simulation to estimate the adult form of the Taung child, and directly compare its morphology to that of other fossil hominins. Specimens were represented by 50 three-dimensional landmarks superimposed by generalized Procrustes analysis. The simulation process applied developmental trajectories from extant hominine species to the Taung fossil in order to generate its adult form. Despite differences found in the developmental patterns of these modern species, simulations tested on extant juveniles-transforming them into "adults" using trajectories from other species-revealed that these differences have negligible impact on adult morphology. This indicates that morphology already present by occlusion of the first permanent molar is the primary determinant of adult form, thereby supporting use of extant trajectories to estimate the morphology of an extinct species. The simulated Taung adult was then compared to other adult fossils. As these comparisons required assumptions about the pattern and magnitude of developmental change, additional analyses were performed to evaluate these two parameters separately. Results of all analyses overwhelmingly rejected the possibility that the Taung child was a juvenile robust australopith, but were consistent with the hypothesis that the Taung and Sterkfontein fossils are conspecific. Between Sts 5 and Sts 71, the latter is more likely to resemble the adult form of the Taung child.     

Dietary adaptations of South African australopiths: inference from enamel prism attitude

The angle at which enamel prisms approach the wear surface holds information with regard to the stiffness of the tissue, as well as its wear resistance. Hence, analyses of prism orientation may shed light on questions of whether the thick enamel in hominins has evolved to confer stiffness or wear resistance to the teeth and may thus inform about the diet and behavioural ecology of these species. This was explored for Paranthropus robustus and Australopithecus africanus, whereby a distinction was made between prisms at the Phase I and Phase II facets. The results were compared with those obtained for Theropithecus, Macaca, and Potamochoerus for whom behavioural and/or experimental data are available, and were interpreted against simple mechanical principles. The South African hominins differ significantly in their relationships between wear facets and prism angulations. Teeth of P. robustus are better adapted to more vertical loads during mastication (Phase I), whereas those of A. africanus are better adapted to cope with more laterally-directed loads (Phase II) commonly associated with roll-crush and mastication. Overall, teeth of P. robustus appear stiffer, while those of A. africanus seem more wear resistant.     

Molar microwear textures and the diets of Australopithecus anamensis and Australopithecus afarensis

Many researchers have suggested that Australopithecus anamensis and Australopithecus afarensis were among the earliest hominins to have diets that included hard, brittle items. Here we examine dental microwear textures of these hominins for evidence of this. The molars of three Au. anamensis and 19 Au. afarensis specimens examined preserve unobscured antemortem microwear. Microwear textures of these individuals closely resemble those of Paranthropus boisei, having lower complexity values than Australopithecus africanus and especially Paranthropus robustus. The microwear texture complexity values for Au. anamensis and Au. afarensis are similar to those of the grass-eating Theropithecus gelada and folivorous Alouatta palliata and Trachypithecus cristatus. This implies that these Au. anamensis and Au. afarensis individuals did not have diets dominated by hard, brittle foods shortly before their deaths. On the other hand, microwear texture anisotropy values for these taxa are lower on average than those of Theropithecus, Alouatta or Trachypithecus. This suggests that the fossil taxa did not have diets dominated by tough foods either, or if they did that directions of tooth–tooth movement were less constrained than in higher cusped and sharper crested extant primate grass eaters and folivores.     

Evolutionary Development in Australopithecus africanus

Evolutionary developmental biology is quickly transforming our understanding of how lineages evolve through the modification of ontogenetic processes. Yet, while great strides have been made in the study of neontological forms, it is much more difficult to apply the principles of evo-devo to the miserly fossil record. Because fossils are static entities, we as researchers can only infer evolution and development by drawing connections between them. The choices of how we join specimens together??juveniles to adults to study ontogeny, taxon to taxon to study evolution??can dramatically affect our results. Here, I examine paedomorphism in the fossil hominin species Australopithecus africanus. Using extant African apes as proxies for ancestral hominin morphology, I demonstrate that Sts 71 is most similar to a sub-adult African ape, suggesting that A. africanus is paedomorphic relative to the presumed ancestral form. I then plot ontogenetic size and shape in extant great apes, humans, and A. africanus in order to assess patterns of ontogenetic allometry. Results indicate that ontogenetic allometry in A. africanus, subsequent to M1 occlusion is similar to that in modern humans and bonobos; gorillas, chimpanzees, and orangutans share a different pattern of size-shape relationship. Combined with results from the analysis of paedomorphism plus knowledge about the developmental chronologies of this group, these findings suggest that paedomorphism in A. africanus arises relatively early in ontogeny.