Once upon a time, a long time ago when I was a high school student, I remember being taught about human evolution as a fairly linear, straightforward narrative. OK, there were those 'robust' australopiths (aka Paranthropus) on a dead-end side branch, but otherwise species followed species - beginning around 14 million years ago with Ramapithecus (or Sivapithecus) - until you got to us. I don't remember being taught much about 'new' finds, either, although there must have been some; it all came across as pretty much done & dusted. In other words, we didn't learn much about the nature of science, either; there was no sense that new discoveries could overturn existing understandings.
Now, of course, things have changed. A lot. We now know that Ramapithecus wasn't a hominin; that the one-day-to-be-human line branched from that of the great apes only 5-6 million years ago; and that modern humans emerged about 200,000 years before the present day. Neandertals have been rehabilitated from the brutish, shuffling apemen that they were first portrayed as. And our family tree, much bushier than when I was taught about it, has become even more complex as finds such as Homo floresiensis are slotted into it.
One useful tool for deciphering our past is molecular biology. In March last year I wrote about evidence of another possible branch in our tree, following the discovery of a hominin finger bone in Denisova Cave, Siberia. At the time an analysis of mtDNA from the fossil suggested that its owner belonged to a lineage that diverged from the branch leading to sapiens & neandertalensis about a million years ago. Now in a paper published in December 2010, David Reich & his colleagues have presented the results of an analysis of nuclear DNA from that same fossil, plus an analysis of the morphology & mtDNA of a tooth from the same cave that appears to confirm the separate evolutionary trajectory of this particular hominin population. (The tooth, although quite large, does fall within the range of tooth sizes for Upper Palaeolithic humans, so in terms of morphology it's not particularly useful as a diagnostic characteristic.)
But the data also show tantalising evidence of at least some degree of genetic intermingling. You may remember that the recent sequencing of the Neandertal genome indicated that there seemed to have been a small amount of gene flow between Neandertals and Eurasian (but not African) populations of Homo sapiens. The Denisova analyses suggest something similar: "that [the Denisovans] contributed 4-6% of [their] genetic material to the genomes of present-day Melanesians" (Reich et al. 2010). As PZ Myers has said, this makes our family tree look more like a braided river, with various branches emerging over time but with some degree of gene flow occurring between those branches from time to time.
However, we do need to remember that the analysis by Reich et al. is based on the remains of only two individuals, with very little in the way of morphological data with which to triangulate their findings. So for the moment the possible position of the Denisovan individuals on our family tree (or stream) is hypothetical, albeit a fascinating hypothesis.
D.Reich, R.E.Green, M.Kircher, J.Krause, N.Patterson, E.Y.Durand, B.Viola, A.W.Briggs, U.Stenzel, P.L.F.Johnson, T.Maricic, J.M.Good, T.Marques-Bonet, C.Alkan, Q.Fu, S.Mallick, H.Li, M.Meyer, E.E.Eichler, M.Stoneking, M.Richards, S.Talamo, M.V.Shunkov, A.P.Derevianko, J-J.Hublin, J.Kelso, M.Slatkin & S.Paabo (2010) Genetic history of an archaic hominin group from Denisova Cave in Siberia. Nature 468: 1053 - 1060 doi: 10.1038/nature09710