Can somebody find these hobbits a mommy?

4 minute read

More on the evolving story of the hobbit remains in this story ( Like most custody battles, the fight over these little guys is burning a lot of people. I don't really have a comment, it's just interesting the way the wind is blowing.

Notice at the bottom that a sample has been sent to the Max Planck Institute for DNA recovery.

In light of recent comments about DNA contamination, I'm now doubtful this will achieve a meaningful result. Consider the possibilities:

  1. The hobbits have only modern human sequences. If this result could be authentically shown to come from endogenous DNA, then this would show that the Brown hypothesis about island dwarfism was wrong, and this indeed was a pathological, but otherwise unremarkable, human. But this result would almost certainly not be accepted, because the modern human sequences would be argued to be contamination. This is a much bigger problem than was usually acknowledged a few years ago, when people maintained that there was some way to authenticate whether sequences were endogenous. Now it is clear that all specimens have substantial modern contamination, and that the vast majority of sequence are of modern origin, even in specimens that actually do preserve ancient sequences. If only modern DNA is found in the hobbits, this will probably be interpreted as contamination whether it is endogenous or not. The only exception to this would be if it could be shown that the hobbit had a sequence that was closely related to some sample found in living Indonesians, but that there was a unique mutation that set it apart, and has never been found. But I am skeptical that even that would be enough; it is very difficult to respond to the argument that a rare sequence might have been present in one of the discoverers, researchers, or others who handled the sample.
  2. The hobbits have racemization and collagen profiles that indicate that DNA should not be preserved. This is probably the most likely result, because Flores seems likely to be a climatic regime that would not facilitate DNA preservation over tens of thousands of years.
  3. The hobbits preserve some sequence slightly divergent from all living people. By slightly divergent, here, I mean around as divergent as Neandertal mtDNA sequences from recent humans. This might seem to be a clear result, but is actually problematic. For example, the Mungo 3 specimen (Adcock et al. 2001b) appears to lie on a clade that is an outgroup to later people, along with a nuclear mtDNA insertion (called a numt) on chromosome 11. But this kind of interpretation faces a lot of potential problems. For example, Cooper and colleagues (2001) argued that the Mungo 3 sequence may actually lie on the human clade, and is associated with the chromosome 11 numt by homoplasy (Adcock and colleagues (2001a) reply to this argument, but it isn't settled). The same criticism has been made of the Neandertal sequences by Gutierrez and colleagues (2002): namely that these actually lie inside the human clade. It would appear that the degree of parallelism among mtDNA control region sequences is so high that most ancient human specimens will suffer from this problem. This is above and beyond the possibility that an ancient sequence may look unique purely because of imperfect preservation and resulting spurious genetic changes (Caldararo and Gabow 2000). And no close genetic difference can escape the criticism that recent human mtDNA has been subject to recent positive selection. If this were true, then a wide variety of genetic sequences may have been common within the human population before the selective sweep, including the Neandertal sequence, the Mungo 3 sequence, the chromosome 11 numt, and any putative Liang Bua sequence. In short, a closely related sequence to humans would not settle the issue of Homo floresiensis, even if it were basal to the clade including all living humans.
  4. The hobbits preserve some sequence radically divergent from those of all living people. I view this as a relatively unlikely result, but would be expected if indeed these hominids descended from some very early human or australopithecine lineage. In this event, the divergence between Liang Bua and humans would be greater than (and perhaps substantially greater than) the difference between humans and Neandertals. The date of such a divergence is very likely to be between 1 million and 2.3 million years ago, considering that human presence on Flores began by around 900,000 years ago (Morwood et al. 1998), and any extra-African excursions by early Homo or late Australopithecus probably occurred after the origins of stone tool manufacture. The date might therefore be substantially informative about the relationships of early humans and australopithecines, depending on what it turns out to be. But again, I regard this result as by far the least likely, and I think it is much more likely that the DNA survey will be entirely uninformative about the relationships of the sampled specimen.


Adcock GJ, Dennis ES, Easteal S, Huttley GA, Jermiin LS, Peacock WJ, Thorne A. 2001a. Human origins and ancient human DNA. Science 292:1655-1666. Science

Adcock GJ, Dennis ES, Easteal S, Huttley GA, Jermlin LS, Peacock WJ, Thorne A. 2001b. Mitochondrial DNA sequences in ancient Australians: Implications for modern human origins. Proc Natl Acad Sci U S A 98:537-542.

Brown P, Sutikna T, Morwood MJ, Soejono RP, Jatmiko, Saptomo EW, Due RA. 2004. A new small-bodied hominin from the Late Pleistocene of Flores, Indonesia. Nature 431:1055-1061.

Caldararo N, Gabow S. 2000. Mitochondrial DNA analysis and the place of Neandertals in Homo. Ancient Biomol 3:135-158.

Cooper A, Rambaut A, Macaulay V, Willerslev E, Hansen AJ, Stringer C. 2001. Human origins and ancient human DNA. Science 292:1655-1666. Science

Gutiérrez G, Sánchez D, Marín A. 2002. A reanalysis of the ancient mitochondrial DNA sequences recovered from Neandertal bones. Mol Biol Evol 19:1359-1366.

Morwood MJ, O'Sullivan PB, Aziz F, Raza A. 1998. Fission-track ages of stone tools and fossils on the east Indonesian island of flores. Nature 392:173-176. Nature