Portal de Periódicos da CAPES

Primate evolution — in and out of Africa

1999; Elsevier BV; Volume: 9; Issue: 15 Linguagem: Inglês

10.1016/s0960-9822(99)80350-9

1879-0445

Salvador Moyà‐Solà, Meike Köhler, David M. Alba, Caro‐Beth Stewart, Todd R. Disotell,

Bat Biology and Ecology Studies

Stewart and Disotell [[1]Stewart CB Disotell TR Primate evolution – in and out of Africa.Curr Biol. 1998; 8: R582-R588Abstract Full Text Full Text PDF PubMed Google Scholar] construct a scenario for the evolutionary history of catarrhine primates that they believe to be “robust about the exact phylogenetic positions of the Eurasian hominoid fossils”, and parsimonious in that it requires a minimum number of dispersal events. The authors present a synthesis of the known molecular phylogeny of the living species and parsimony analyses of fossils, and map the biogeographical locations of the living and fossil species onto this phylogeny. Stewart and Disotell conclude that “the lineage leading to the living hominoids dispersed out of Africa about twenty million years ago, and that the common ancestor of the living African apes, including humans, migrated back into Africa from Eurasia within about the past ten million years.” We think that multi-disciplinary analyses involving data from different fields such as molecular biology, biogeography and paleontology are useful, and even necessary, to reliably reconstruct the evolutionary history of a phylogenetic entity, and we applaud this approach. Nevertheless, it is a standard procedure in scientific analyses to verify whether the hypothesis is consistent with the known data. In this case, the proposed scenario B is not consistent with the fossil record. In Eurasia, Miocene faunas are documented from hundreds of rich localities, and the first appearances of the different primate groups are well dated [2Andrews P Harrison T Delson E Bernor RL Marin L Distribution and biochronology of European and Southwest Asian Miocene catarrhines.in: Bernor RL Fahlbusch V Mittman H-W The Evolution of Western Eurasian Neogene Mammal Faunas. Plenum Press, New York1997: 168-207Google Scholar, 3Bernor RL Flynn LJ Harrison T Hussain ST Kelley J Dionysopithecus from southern Pakistan and the biochronology and biogeography of early Eurasian catarrhines.J Hum Evol. 1998; 17: 339-358Crossref Scopus (38) Google Scholar]. No catarrhine primate is known from any site older than 16 million years ago, despite faunal, floral and sedimentological testimonies of environmental conditions suitable for hominoids. The oldest Asian catarrhine (16.1 million years ago) is Dionysopithecus, a hominoid without clear affinities to either hylobatids, orangutans or African apes [[4]Barry CB Jacobs LL Kelley J An Early Middle Miocene catarrhine from Pakistan with comments on the dispersal of catarrhines into Eurasia.J Hum Evol. 1998; 15: 501-508Crossref Scopus (20) Google Scholar]. The oldest European catarrhines, the non-hominoid pliopithecids and the primitive hominoid Griphopithecus, first appeared between 15.5 and 15 million years ago [[2]Andrews P Harrison T Delson E Bernor RL Marin L Distribution and biochronology of European and Southwest Asian Miocene catarrhines.in: Bernor RL Fahlbusch V Mittman H-W The Evolution of Western Eurasian Neogene Mammal Faunas. Plenum Press, New York1997: 168-207Google Scholar]. Thus, an ‘out of Africa’ dispersal of hominoids at 20 million years ago must be discarded. Taking into account the age of the lesser ape / great ape dichotomy provided by molecular data as roughly 18 million years ago, it seems more reasonable to assume two dispersals into Eurasia, one of the hylobatids at about 16 million years ago and another of the Eurasian great ape ancestor between 13 and 12 million years ago. In fact, a third event took place, when Griphopithecus — a descendant of Kenyapithecus and probably not related to the younger Eurasian apes — dispersed into Eurasia about 15.5 million years ago. The gap in the African fossil record between 12 and 6.5 million years ago does not indicate a putative extinction of the African hominoid lineages, but is likely to be an artifact. Extant hominoids and almost all fossil hominoids have a pantropical distribution; fossils from tropical environments are scarce. Unlike the large record from Eurasian localities, the African Upper Miocene–Lower Pliocene record is poor, most likely because of the difficulties of sampling in the areas where these fossil hominoids probably lived. Thus, the survival of one or more hominoid lineages in Africa during the Upper Miocene is likely and an African origin of the African ape and human clade must be considered. Given that the well-sampled Eurasian fossil record is not consistent with a Eurasian origin of the lesser and great ape lineages, scenario B proposed by Stewart and Disotell must be ruled out, as it relies on this premise. We propose returning to the two alternative scenarios being discussed by some paleontologists, which are based on reliable paleontological data suggesting an African origin for both the hylobatids and the extant great apes. Scenario 1 suggests a branching of the Eurasian apes with the orangutan clade and an African origin for the African ape and human clade [5Moyà-Solà S Köhler M Recent discoveries of Dryopithecus shed new light on evolution of great apes.Nature. 1993; 365: 543-545Crossref PubMed Scopus (77) Google Scholar, 6Moyà-Solà S Köhler M New partial cranium of Dryopithecus Lartet, 1863 (Hominoidea, Primates) from the upper Miocene of Can Llobateres, Barcelona, Spain.J Hum Evol. 1995; 29: 101-139Crossref Scopus (85) Google Scholar, 7Begun DR Ward CV Rose MD Events in hominoid evolution.in: Begun DR Ward CV Rose MD Function, Phylogeny and Fossils: Miocene Hominoid Evolution and Adaptation. Plenum Press, New York1997: 389-415Crossref Google Scholar, 8Schwartz JH Lufengpithecus and its potential relationship to an orang-utan clade.J Hum Evol. 1990; 19: 591-605Crossref Scopus (35) Google Scholar, 9Miyamoto MM Young TS Primate evolution – in and out of Africa.Curr Biol. 1998; 8: R745-R746Abstract Full Text Full Text PDF PubMed Google Scholar] (see Figure 1a). Scenario 2 suggests a Eurasian origin for the African ape ancestor and assumes an additional ‘back into Africa’ migration [[7]Begun DR Ward CV Rose MD Events in hominoid evolution.in: Begun DR Ward CV Rose MD Function, Phylogeny and Fossils: Miocene Hominoid Evolution and Adaptation. Plenum Press, New York1997: 389-415Crossref Google Scholar] (see Figure 1b). Both scenarios are consistent with the molecular data but differ in the selection, interpretation and evaluation of characters for reconstructing the phylogenetic relationships, as well as in the resulting dispersal patterns of the fossil apes. But computer-assisted analyses, even when based on the controversial [6Moyà-Solà S Köhler M New partial cranium of Dryopithecus Lartet, 1863 (Hominoidea, Primates) from the upper Miocene of Can Llobateres, Barcelona, Spain.J Hum Evol. 1995; 29: 101-139Crossref Scopus (85) Google Scholar, 7Begun DR Ward CV Rose MD Events in hominoid evolution.in: Begun DR Ward CV Rose MD Function, Phylogeny and Fossils: Miocene Hominoid Evolution and Adaptation. Plenum Press, New York1997: 389-415Crossref Google Scholar, 10Köhler M Moyà-Solà S Fossil muzzles and other puzzles.Nature. 1997; 388: 327-328Crossref PubMed Scopus (4) Google Scholar] data-matrix of Begun et al. [[7]Begun DR Ward CV Rose MD Events in hominoid evolution.in: Begun DR Ward CV Rose MD Function, Phylogeny and Fossils: Miocene Hominoid Evolution and Adaptation. Plenum Press, New York1997: 389-415Crossref Google Scholar] reveal that both phylogenies differ in less than 2% step length (scenario 1 is only six steps longer than scenario 2). Thus, both are considered highly parsimonious, even by those who prefer scenario 2 [[7]Begun DR Ward CV Rose MD Events in hominoid evolution.in: Begun DR Ward CV Rose MD Function, Phylogeny and Fossils: Miocene Hominoid Evolution and Adaptation. Plenum Press, New York1997: 389-415Crossref Google Scholar]. In searching for the biogeographic origin of African apes, however, Stewart and Disotell ignore scenario 1 and consider only scenario 2, which relies on a controversial hominoid extinction event in Africa and where the ‘back into Africa’ migration requires an additional dispersal event. Scenario 1 — which is consistent with the fossil record, acceptable from computer-assisted cladistic analyses and requires a minimum number of dispersal events — is likely to be the most parsimonious scenario to date. We wonder why Stewart and Disotell refused to consider this hypothesis, which is widely discussed in the current literature. This work was supported by the Wenner Gren Foundation of Anthropological Research (M.K and S.M.S.) and a predoctoral fellowship (FI) from the Generalitat de Catalunya (D.M.A.). Our dispersal scenario for the hominoids [[1]Stewart CB Disotell TR Primate evolution – in and out of Africa.Curr Biol. 1998; 8: R582-R588Abstract Full Text Full Text PDF PubMed Google Scholar] — unlike that of Moyà-Solà and colleagues — is built upon analysis of known fossils and not upon arguments concerning absence of fossils. Indeed, our suggestion that the early African hominoids such as Kenyapithecus went extinct, rather than giving rise to the living African hominoids, is the logical inference resulting from the topology of the most-parsimonious tree of the living and fossil hominoids (see Figure 2 in [[1]Stewart CB Disotell TR Primate evolution – in and out of Africa.Curr Biol. 1998; 8: R582-R588Abstract Full Text Full Text PDF PubMed Google Scholar] and Figure 2c below). This inference is consistent with the well-known gap in the African hominoid fossil record between about 12 million years ago (the last appearance of Kenyapithecus) and 9.5 million years ago (the first appearance of Samburupithecus), but this lack of fossil evidence was not the foundation of our reasoning. We also take issue with the claim that the two scenarios presented in Figure 1a are “widely discussed in the current literature.” Indeed, these phylogenetic hypotheses are not explicitly presented or tested even in the authors’ cited papers [5Moyà-Solà S Köhler M Recent discoveries of Dryopithecus shed new light on evolution of great apes.Nature. 1993; 365: 543-545Crossref PubMed Scopus (77) Google Scholar, 6Moyà-Solà S Köhler M New partial cranium of Dryopithecus Lartet, 1863 (Hominoidea, Primates) from the upper Miocene of Can Llobateres, Barcelona, Spain.J Hum Evol. 1995; 29: 101-139Crossref Scopus (85) Google Scholar]. In [[5]Moyà-Solà S Köhler M Recent discoveries of Dryopithecus shed new light on evolution of great apes.Nature. 1993; 365: 543-545Crossref PubMed Scopus (77) Google Scholar], for example, the authors merely present their opinion — without support from explicit data analysis — about the phylogenetic position of Dryopithecus, and do not include Proconsul, Kenyapithecus or Oreopithecus in the trees. Likewise, it is not true that the three scenarios presented in Figure 1 have been tested by “computer-assisted analyses” and found to be essentially equivalent. Indeed, neither of the phylogenies shown in Figure 1a have the same taxa or topologies as those shown in ‘scenario B’ of [[9]Miyamoto MM Young TS Primate evolution – in and out of Africa.Curr Biol. 1998; 8: R745-R746Abstract Full Text Full Text PDF PubMed Google Scholar] or ‘hypothesis E’ of [[7]Begun DR Ward CV Rose MD Events in hominoid evolution.in: Begun DR Ward CV Rose MD Function, Phylogeny and Fossils: Miocene Hominoid Evolution and Adaptation. Plenum Press, New York1997: 389-415Crossref Google Scholar]. Although Moyà-Solà and Köhler [[6]Moyà-Solà S Köhler M New partial cranium of Dryopithecus Lartet, 1863 (Hominoidea, Primates) from the upper Miocene of Can Llobateres, Barcelona, Spain.J Hum Evol. 1995; 29: 101-139Crossref Scopus (85) Google Scholar] did compile 20 characters for 11 hominoid taxa, only 18 of the characters were informative for parsimony analysis, and 4 of the taxa lacked 5–12 of these informative characters. Moreover, these 11 taxa only partially overlap those shown in Figure 1. Specifically, in the previous study [[6]Moyà-Solà S Köhler M New partial cranium of Dryopithecus Lartet, 1863 (Hominoidea, Primates) from the upper Miocene of Can Llobateres, Barcelona, Spain.J Hum Evol. 1995; 29: 101-139Crossref Scopus (85) Google Scholar] the authors included Afropithecus (which they omit here) but did not include Griphopithecus, Oreopithecus or Australopithecus. Such confounding factors make the authors’ present numerical comparisons of the various phylogenetic hypotheses completely inapplicable. In Figure 2, we present appropriate statistical comparisons between the three phylogenetic hypotheses presented in Figure 1. Following Miyamoto and Young [[9]Miyamoto MM Young TS Primate evolution – in and out of Africa.Curr Biol. 1998; 8: R745-R746Abstract Full Text Full Text PDF PubMed Google Scholar], we used PAUP∗[[15]Swofford DL PAUP∗ 4.0: Phylogenetic Analysis Using Parsimony and Other Methods., MA. Sinauer, Saunderland1998Google Scholar] to infer the most-parsimonious tree from the 240-character data set [[7]Begun DR Ward CV Rose MD Events in hominoid evolution.in: Begun DR Ward CV Rose MD Function, Phylogeny and Fossils: Miocene Hominoid Evolution and Adaptation. Plenum Press, New York1997: 389-415Crossref Google Scholar] and to compare the trees by statistical tests. The most-parsimonious tree found (length = 389 steps) is shown in Figure 2c; it has the same topology as ‘scenario 2’ shown in Figure 1b. The tree that places Oreopithecus as a sister-taxon to Dryopithecus (Figure 1a and Figure 2a; length = 424), and therefore hypothesizes a monophyletic clade of Eurasian great apes, was ruled out (p < 0.0001). Likewise, the tree that places Oreopithecus basal to the great apes (Figure 1a and Figure 2b; length = 414) was ruled out (p = 0.0009). Thus, the most-parsimonious tree is statistically significantly better than the two competing hypotheses presented in Figure 1a. Logically, this parsimony tree requires only two dispersal events for the hominoids: one out of Africa by a common ancestor of the lesser and great apes, and one back into Africa by a common ancestor of the African apes and humans. Importantly, all three trees strongly suggest that the ancestor of the great apes dispersed from Africa to Eurasia much earlier than 13 million years ago, as Sivapithecus fossils date to this age and several of the Eurasian ape divergence points predate the Sivapithecus–Pongo split on these trees. So, although it is theoretically possible that the gibbon and great ape lineages dispersed out of Africa separately, the most parsimonious explanation is that their common ancestor did so before about 18 million years ago. A lack of known hominoid-like fossils in Eurasia dating to this period is irrelevant to this argument. First, the fact that hominoid-like fossils have not yet been discovered in Eurasian fossil beds from this period does not mean that the ancestral species did not exist on this land mass. Second, the ancestral catarrhine that dispersed from Africa to Eurasia may have been more monkey-like than ape-like in many features, as the hominoid lineage clearly evolved from a monkey-like ancestor. Indeed, recent analyses [[14]Harrison T Yumin G Taxonomy and phylogenetic relationships of early Miocene catarrhines from Sihong, China.J Hum Evol. 1999; 37: 225-277Crossref PubMed Scopus (60) Google Scholar] suggest that the catarrhine fossil genera Dionysopithecus and Platodontopithecus are actually pliopithecids that had arrived in China by about 18 million years ago, having dispersed out of Africa during a major intercontinental faunal exchange that occurred about 18–22 million years ago. We would like to make the heterodox suggestion that these Asian catarrhines may represent the elusive common ancestors of the living apes. Testing such hypotheses will require phylogenetic and statistical analyses of the fossils from early catarrhines. S Moyà-Solà, M Köhler and DM Alba, Institut de Paleontologia M. Crusafont, Escola Industrial 23, 08201 Sabadell, Barcelona, Spain. CB Stewart, Department of Biological Sciences, University of Albany, State University of New York, 1400 Washington Avenue, Albany, New York 12222, USA. TR Disotell, Department of Anthropology, New York University, 25 Waverly Place, New York, New York 10003, USA.