Research Center
The Bookstore
Origins Links
Origins News

Help Develop Site
the HOIT

Jeff Rix
780 Stonegate Dr.
Wexford, PA 15090

Copyright© 1995-96
Origins of Humankind
All Rights Reserved
Human Origins News
See Milford H. Wolpoff's new book Race and Human Evolution.

No "Homo erectus" at Ngandong
Submitted by:
Milford H. Wolpoff
Department of Anthropology
University of Michigan
Ann Arbor, MI 48109-1382, USA
313 475-3291
January 4, 1997
Related Graphic

Human remains from Ngandong, and the Sambungmachan skullcap, were described as "Homo erectus"1 because of important similarities that point to a specific relationship between them and the earlier Kabuh Indonesians2 also assigned to "Homo erectus". This relationship is almost certainly descent3 . One can argue that neither taxonomic assignment is appropriate because there is ample justification to question whether "Homo erectus" is a valid taxon or if it should be subsumed within Homo sapiens4 . But even if we assume its validity there are a number of reasons why Ngandong and Sambungmachan must be regarded as Homo sapiens and not "Homo erectus":

  1. Ngandong and Sambungmachan show notable similarities to Late Pleistocene Native Australians;
  2. The Kabuh Indonesians are among the ancestors of Holocene and living Native Australians and Ngandong and Sambungmachan are temporally5 and anatomically intermediate between them. If they were both "Homo erectus" and dated as recently as suggested, it would suggest the interpretation of polygenism - different populations became Homo sapiens at different times - and this is unacceptable;
  3. The evolutionary trends in a Kabuh to Ngandong lineage are the same as trends in other evolving Homo sapiens populations and it is unlikely these changes reflect parallelisms between species.
These points are detailed below.

1. Ngandong is an especially well-studied sample6 and similarities between it and recent and living Native Australians are often described7 . Yet Swisher and colleagues claim recent and living Native Australians do not descend from Ngandong5 and that earlier Australian fossils such as WLH 50 are unrelated as well. They reject the contention that the Ngandong sample is Homo sapiens and argue against any relationship between Ngandong and Late Pleistocene Australians that does not involve cross-species gene flow. They assert the features shared by Ngandong and Australian fossils are plesiomorphic because they "are equally prevalent in fossil hominids from North Africa", citing a source8 for this conclusion that, however, does not address the issue of how Ngandong and fossil Native Aboriginal Australian crania might be related. In fact, a number of features link Ngandong with Australian fossil remains that are not generally found in Africans. A comparison of WLH 50 with Ngandong and Ngaloba (Laetoli 18), the most archaic-appearing of the African fossils of similar age and an alternative ancestor for WLH 50 according to the "Out of Africa" theory, found 12 features which linked WLH 50 with the Ngandong sample9 to the exclusion of Ngaloba. Not one feature could be found that uniquely linked WLH 50 and Ngaloba to the exclusion of Ngandong10 . It is surprising to find so many similarities in the absence of faces and very unlikely that so detailed a series of resemblances could be due to chance. Comparisons of WLH 50 with Ngandong and Qafzeh, another potential "African ancestor"11 for the Late Pleistocene Australian colonizers, indicate the same conclusion12 .

These similarities are especially troublesome for the "Homo erectus" attribution because even the earliest of the Late Pleistocene Australians are behaviorally modern. The physical remains of Australia’s earliest inhabitants are yet to be discovered. Yet their modernity is evident. They initiated continued crossings of formidable water barriers, too wide to see across, which implies complex technological and organizational skills13 .

2. Numerous similarities link the much earlier Indonesian hominids from Kabuh to recent and living Native Indigenous Australians14 and we can expect there were intermediates2 between some of these 750,000 year old hominids and some living populations. The claim1 that Ngandong cannot be such an intermediate because "Homo erectus" from Ngandong overlaps in time with Homo sapiens from Australia, is problematic. There is a problem, not only because no dated specimens have yet unequivocally demonstrated such an overlap15 , but because Ngandong could reflect a transitional anatomy for two different reasons: (1) because it is a direct ancestor, or (2) as a mutual descendent of a common ancestor with Native Australians. Both could be correct, but if either is correct Ngandong cannot be "Homo erectus".

The new dates cannot render the transitional interpretation "no longer chronologically plausible" as is claimed, unless one assumes that the Australians were a unique human line that became isolated from the rest of the world once Australia was colonized. Otherwise Indonesia could have continued contributing colonists descended from Ngandong, which would thereby be intermediate between Kabuh and some living Australians16 . What the dates render implausible is the possibility that Ngandong can be late, and be "Homo erectus" and be among the ancestors of Native Australians, because it would mean that "Homo erectus" in this region became Homo sapiens later than in other parts of the world - an unacceptable position17 . Since Swisher and colleagues insist on the taxonomy and the date, only the transitional nature of the sample is left for them to question. But I believe the weight of the data suggest the problem, and its solution, lies in the taxonomy.

3. Ngandong differs from its Kabuh ancestors and approaches the anatomy Homo sapiens elsewhere in many ways. For instance it has significant supraorbital torus reduction (comparing like sexes, the torus is smaller than the Kabuh specimens, and a depression over the nose results in its distinct separation into right and left sides). The frontal bone is markedly broader, especially across the frontal lobes (behind the orbits, where the postorbital construction is less). The articular eminence for the mandible is projecting and well defined. The occipital plane of the occipital bone is markedly expanded while the nuchal muscle attachment area is decreased by some 30%18 . But most importantly the relationship can be seen in brain size. The Ngandong sample has a brain size considerably expanded over the Kabuh hominids, not just within the Native Aboriginal Australian range but closely approaching the mean.

Female average Male average Brain Volumes19 in
cubic centimeters
Indonesian Kabuh 875 (n=5) 1032 (n=2)
Indonesian Ngandong 1093 (n=2) 1177 (n=4)
Native Australian 1119 (n=22) 1239 (n=51)

These common changes would have to be explained by parallelism if "Homo erectus" persisted on Java while Homo sapiens was evolving the same way in other places. One complex parallelism is possible but combined they pose a statistical impossibility that this interpretation cannot overcome.

In the assessment of Ngandong as "Homo erectus", regional features have been confused with traits that are mistakenly taken to indicate evolutionary grade. For instance, thickened supraorbital tori divided from a flat frontal squama by a weak sulcus characterize many of the Kabuh hominids and Ngandong specimens. However this does not mean Ngandong is plesiomorphic, because the same particular frontal bone anatomy is found continuously throughout the Australasian fossil record20 and is different from the frontal bone anatomy common in other areas. Recent Native Aboriginal Australians with this anatomy are not more plesiomorphic than other recent human populations. Explaining anatomical similarities such as these by cross-species gene flow raises the specter that some human populations can be interpreted to differ from others because they have more genes from an extinct primitive human species.

I do not believe this is correct. WLH 50 and other Australian fossils are modern humans. The bases for this assertion are both their resemblance to more recent Native Australians21 and where they died. The numerous similarities of these modern human crania to the Ngandong crania, in turn, combine with other comparisons of Ngandong and Native aboriginal Australians to suggest that the Ngandong specimens are directly ancestral to the modern humans of this region in the same sense that any of the regional Pleistocene predecessors of modern populations are; that is, as one of several ancestral populations. Ngandong has a similar relationship to later Australians that, for instance, the Klasies remains have to Pleistocene South Africans or Skhul and Qafzeh remains have to later Levantines. Ngandong, by these criteria, cannot be "Homo erectus".

  1. C.C. Swisher III, W.J. Rink, S.C. Antón, H.P. Schwarcz, G.H. Curtis, A. Suprijo, and Widiasmoro, Science 274, 1870 (1996).
  2. But not as unique ancestors. As Weidenreich put it (The skull of Sinanthropus pekinensis: A comparative study of a primitive hominid skull. Palaeontologia Sinica, New Series D(10), pp. 249-250 (1943)): “at least one line leads from Pithecanthropus and Homo soloensis to the Australian aborigines of today. This does not mean … all the Australians of today can be traced back to Pithecanthropus or that they are the sole descendants of the Pithecanthropus-Homo soloensis line”.
  3. F. Weidenreich, Morphology of Solo man. Anthropological Papers of the American Museum of Natural History 43(3), 205 (1951).
  4. E. Aguirre, in 100 years of Pithecanthropus: The Homo erectus problem, J.L. Franzen, Ed. Courier Forschungsinstitut Senckenberg 171:333 (1994); H. Hemmer Current Anthropology 10(2-3), 179 (1969); J. Jelínek, Recent Advances in Primatology 3, 419 (1978); R.E. Leakey, in Human Origins, J.R. Durant, Ed. (Clarendon Press, Oxford, 1989), p. 53-62; J. Robinson, Evolutionary Biology 1, 69 (1967); M.H. Wolpoff, A.G. Thorne, J. Jelínek, and Zhang Yinyun, in 100 years of Pithecanthropus: The Homo erectus problem, J.L. Franzen, Ed. Courier Forschungsinstitut Senckenberg 171, 341 (1994)
  5. If the reported dates are correct (although see Gibbons 1996, Science 274, 1841 (1996)) the earliest Australian colonizers are earlier than Ngandong, but the numerous Terminal Pleistocene and Holocene remains from sites such as Kow Swamp and Coobol Crossing remain later.
  6. F. Weidenreich3; T. Jacob, Some Problems Pertaining to the Racial History of the Indonesian Region. (Drukkerij Neerlandia, Utrecht, 1967); A.P. Santa Luca, The Ngandong Fossil Hominids. Yale University Publications in Anthropology, 78 (Yale University Press, New Haven, 1980).
  7. A.G. Thorne and M.H. Wolpoff (American Journal of Physical Anthropology 55, 337 (1981). S.L. Larnach and N.W.G. Macintosh (in Grafton Elliot Smith: The Man and his Work, A.P. Elkin and N.W.G. Macintosh, Eds. (University of Sydney Press, Sydney, 1974), p. 95) compared a number of Australian and New Guinea crania with Europeans and Africans, scoring them for the 18 characters that Weidenreich had thought were unique for the Ngandong hominids. 6 of these were absent in the modern samples, but 9 of the 12 other features could be found and attained their highest frequencies in the Australian and New Guinea natives: the large rounded zygomatic trigone; absence of a supraorbital sulcus; suprameatal tegmen; transverse squamo-tympanic fissure; angling of the petrous to tympanic in the petro-tympanic axis; lambdoidal protuberance; marked ridge-shaped occipital torus; external occipital crest emerging from the occipital torus; marked supratoral sulcus on the occiput.
  8. M.M. Lahr, Journal of Human Evolution 26(1), 23 (1994). This reference is badly flawed in many of the comparisons it does present. In comparing Kabuh, fossil North Africans, and modern Indigenous Aboriginal Australians, certain critical features are inaccurately represented. In some cases general robustness is confused with anatomical details (for instance, the gross size of supraorbital tori is compared across populations instead of the detailed anatomical form) while in other cases the anatomy compared is incorrectly defined (for instance flattened foreheads are said to link the Australasian samples, but in the Lahr study forehead flattening is confused with forehead elevation).
  9. These are the placement of maximum cranial breadth across the supramastoid crests, absence of a parietal boss, pronounced angular torus, marked mastoid height relative to its basal width, marked sagittal keel on the anterior half of the parietals, smooth sagittal region at bregma, elongated frontal bone relative to vault length, frontal bone flattening in the sagittal plane posterior to the supraorbital region, temporal ridge extending across the entire frontal squama, nuchal torus and supratoral notch, strongly developed inferior pointing external occipital protuberance, occipital bulging along the lambdoidal suture at or near the midline.
  10. D.W. Frayer, M.H. Wolpoff, F.H. Smith, A.G. Thorne, and G.G. Pope, Anthropologist 95(1),14 (1993).
  11. R.G. Klein Evolutionary Anthropology 1(1), 5 (1992).
  12. D.W. Frayer, M.H. Wolpoff, A.G. Thorne, F.H. Smith, and G.G. Pope, American Anthropologist 96(2), 424 (1994), figure 1.
  13. G.G. Pope, Natural History 101(10), 48 (1989); I. Davidson and W. Noble, Archaeology in Oceania 27(3), 113 (1992).
  14. For instance see A.G. Thorne and M.H. Wolpoff (American Journal of Physical Anthropology 55, 337 (1981)) for cranial features and A. Kramer (American Journal of Physical Anthropology 86, 455 (1991)) who found seven traits that both characterize the fragmentary Early Pleistocene Indonesian mandibles and uniquely resemble modern Australians, but specifically differ from Africans
  15. For instance, Native Australians the same age and earlier than Ngandong might be found to resemble Ngandong - a possibility enhanced by WLH 50.
  16. This is the fundamental claim of Multiregional evolution in any event, excepting extinctions ancient populations have multiple descendants and modern populations have multiple ancestors (M.H. Wolpoff and R. Caspari, Race and Human Evolution, Simon and Schuster, New York, 1997).
  17. Th. Dobzhansky, Scientific American 208(2):169 (1963).
  18. In Early and Early Middle Pleistocene humans everywhere the nuchal attachment area makes up the major part of the bone.
  19. M.H. Wolpoff, Human Evolution, 1997 Edition (McGraw-Hill, New York, 1997).
  20. P.J. Habgood , in Continuity or Replacement? Controversies in Homo sapiens Evolution, G. Bräuer and F.H. Smith, Eds. (Balkema, Rotterdam, 1992), p. 273.
  21. WLH 50 has very specific similarities to several of the Holocene Australian specimens from Kow Swamp and Coobol Crossing19.
  22. After C.S. Larsen, R.M. Matter, and D.L. Gebo, Human Origins. The Fossil Record, Second Edition. (Waveland Press, Prospect Heights, 1991); Müller -Beck, Urgesichte der Menschenheit, (Hohlhammer, Stuttgardt, 1952); Stringer in Continuity or Replacement? Controversies in Homo sapiens Evolution, G. Bräuer and F.H. Smith, Eds. (Balkema, Rotterdam, 1992), p. 9; and B. Vandermeersch, Les hommes fossiles de Qafzeh (Israël). (Cahiers de Paléoanthropologie, Centre National de la Recherche Sciéntifique, Paris, 1981).