Technical Comments

Comment on “Early Homo at 2.8 Ma from Ledi-Geraru, Afar, Ethiopia”

See allHide authors and affiliations

Science  19 Jun 2015:
Vol. 348, Issue 6241, pp. 1326
DOI: 10.1126/science.aab0591


Villmoare et al. (Reports, 20 March 2015, p. 1352) report on a hominin mandible from the Ledi-Geraru research area, Ethiopia, which they claim to be the earliest known representative of the genus Homo. However, certain measurements and observations for Australopithecus sediba mandibles presented are incorrect or are not included in critical aspects of the study. When correctly used, these data demonstrate that specimen LD 350-1 cannot be unequivocally assigned to the genus Homo.

Specimen LD 350-1 is substantially similar to mandibles of australopiths in dental dimensions, corpus size, and anatomy (Fig. 1). Despite its generally australopith-like morphology, Villmoare et al. (1) attribute LD 350-1 to Homo because of (i) the parallel alveolar and basal corpus margins, with corpus height at P3 only slightly more than at M2; (ii) the posteriorly directed mental foramen; (iii) the posteriorly positioned anterior margin of its ascending ramus; (iv) the nearly vertical buccal walls of its M2 and M3; (v) the buccolingually narrow and relatively salient cusp apices; (vi) the mesially tapered M2 and M3 compared to australopiths; and (vii) its mesiodistally short M3.

Fig. 1 Figures S1A, S2 B, S3 C, and S6D from (1) with A. sediba specimens MH1 and MH2 added to the plots (stars).

The MH1 and MH2 locations are newly added to these plots, except for (D), in which (1) used an incorrect measurement. H. habilis and H. rudolfensis specimens are indicated in each plot by “H” markers, and African H. erectus specimens are indicated by “E” markers, as in (1). The original convex hull plots belong to A. afarensis (red), A. africanus (blue), and Paranthropus boisei (green). (A) The P4 of MH2 is markedly reduced relative to LD 350-1. (B) M2 dimensions do not distinguish A. sediba, A. afarensis, or LD 350-1. (C) In M3 MD length, A. sediba is likewise shorter than nearly all specimens attributed to H. habilis or H. rudolfensis. The convex hull surrounding these early Homo species has been added to show the clear difference between LD 350-1 and these early Homo samples. (D) The correct measurement for corpus height at M2 makes a marked difference to the position of A. sediba in this comparison. MH2 is clearly more “Homo-like” than LD 350-1, and the juvenile MH1 is comparable to other australopiths. The variability of this relation between anterior and posterior corpus height in early Homo also appears to characterize A. sediba.

Parallel alveolar and basal corpus margins are found in the Australopithecus sediba specimen MH2, contrary to Villmoare et al., who provide inaccurate P3 and M2 corpus height measures of 31.1 and 27.0 mm, respectively [table S5 in (1)]. The actual measures, taken from the reassembled original fossil, are 31.0 and 30.5 mm, respectively (Fig. 1), and the nearly parallel nature of the alveolar and basal margins of MH2 is visible in published photographs [figure 1 in (2). Villmoare et al. also suggest that the mandibular corpus in MH1 of A. sediba is deep anteriorly [text S3 in the supplementary materials of (1)], but the specimen is a subadult (3), so corpus depth cannot be reliably established at the level of M2. In addition, several specimens of early Homo show anteriorly divergent alveolar and basal corpus margins, so parallel corpus margins cannot be considered an exclusive indicator of Homo. The mental foramina in MH1 and MH2 of A. sediba are oriented predominantly laterally (not anteriorly in MH2, as inaccurately reported by Villmoare et al.), similar to nearly all specimens of early Homo, with a slight posterior orientation similar if not identical to that of LD 350-1. The origin of the ascending ramus is distinctly posteriorly positioned in MH2 of A. sediba, contrary to the claim of Villmoare et al. that it originates between the M2/M3 [table S6 in (1)]. In fact, the ascending ramus of MH2 takes its origin at about mid-M3, identical to that of LD 350-1, as can be seen in published photographs of MH2 [figure 1 in (2) and figure 1 in (3)]. Therefore, the mandibular corpus does not support LD 350-1’s assignment to Homo.

Dental characters and measurement also do not support such an assignment (Fig. 1). The sloping buccal faces of the molars are different from MH1 of A. sediba but not from MH2. Only the mesial tapering of M2 and M3 and the short mesiodistal length of M3 separate LD 350-1 from A. afarensis, A. africanus, and A. sediba. The definition of Homo cannot hinge on M3 mesiodistal length, because every specimen of H. habilis and H. rudolfensis known in Africa has a longer M3 than LD 350-1, MH1, or MH2. With reference to mesial tapering, as noted by Villmoare et al., there is substantial overlap in mesial:distal breadth ratios between Australopithecus and early Homo, and the unique tapering of LD 350-1 cannot exclusively link it with Homo.

We do not assert that LD 350-1 should be allocated to A. afarensis or to A. africanus, nor do we contend that LD 350-1 belongs in A. sediba, although the evidence cannot reasonably reject these hypotheses. Rather, we would urge caution when assessing the taxonomic affinities of such isolated remains, because at present we cannot be certain what the rest of the dentition, skull, or skeleton of LD 350-1 might have looked like.


View Abstract

Navigate This Article