Homo floresiensis describer Peter Brown has kindly sent me a link to his own website, where he lays out evidence against the claims for recent dental work on the LB1 specimen:
The left first mandibular molar of LB1, Homo floresiensis, is heavily worn. Most of the enamel has been removed from the occlusal surface. The remaining enamel forms a ridge on the buccal and lingual margins, and there is a thin platform of remaining enamel in the disto-lingual quadrant. The softer dentine is somewhat scooped out and has a flat white appearance. There is some adhering sediment on the occlusal surface. Absolutely no evidence of any dental work, temporary filling or anything else. The tooth wear and oral health of LB1 are in all respects typical of older palaeolithic and hunter/gatherer humans, and living apes, and distinct from the mesolithic and more recent human burials in the Holocene layers at Liang Bua.
Brown's discussion includes high resolution photos of the specimen, the 3-D CT reconstruction featured in the Scientific American web story, and CT slices taken through the middle of the left and right teeth. I didn't think the 3-D CT slice was quite right to establish that the tooth was normally worn without question, since it cut through the buccal cusps which are unaltered in any event, but it does show a pulp cavity of normal dimensions for that area.
The slice taken through the centers of the left teeth, although a bit fuzzy (again, characteristic of the CT resolution), is much less equivocal: it shows a normal pulp cavity of equivalent dimensions to the right side and no evidence of alteration or drilling.
That's enough to convince me.
The rest of Brown's description serves to support his experience in examining archaeological teeth, including some photos of worn teeth of various stages. Some of this description will be interesting to readers who may not be as familiar with dental remains (or for that matter to dentists who aren't that familiar with archaeological samples of teeth). I think that these comparisons are sufficient to show that the particular pattern of wear and breakage on the LB1 lower left M1 is a bit odd compared to normal wear. But given that the visible material is in fact dentine (a fact established by the CT), there's nothing else that is outside the scope of either premortem or postdepositional processes. Any single specimen is likely to have idiosyncrasies, and by now it is abundantly obvious that LB1 is no exception to this rule.
UPDATE (2008/04/23): Elizabeth Culotta has a nice story about the tooth online at ScienceNOW.