I've read conflicting reports about the status of Archaeopteryx and its relationship to modern birds. The consensus clearly seems to be that, despite its popular title as the 'first bird' ( a generalisation I hate having to explain to students!), it was not on the line that led to modern Aves. Was it therefore an early enantiornithine, or a basal form that was ancestral to both them and neornithines, or is it simply too difficult to call?
Thanks.

Hi, Andrew.  It depend on what you mean by "the line that led to modern birds".  As you know, evolution does not move in straight lines, but in trees: each ancestor in general begets a whole tree of diverging descendents, so that the chances of our ever finding a true ancestor in the fossil record as pretty remote -- and our chances of identifying it as such even more so.  So when we talk about something being "on the line that led to modern birds", what we really mean "diverged from the line that leads to modern birds, but maybe didn't diverge very much".  (Of course, you and I are on the line that led to modern birds if we go back far enough!)

So the questions about Archaeopteryx are really: *when* did it diverge from the line leading to modern birds (e.g. before or after Velociraptor) and how much did it differ morphologically from the common ancestor.  And those questions are still up for grabs: for example, Mayr (2005) in a description of a new Archaeopteryx specimen, found Archaeopteryx to be basal to troodontids and dromaeosaurs -- so that Velociraptor is more closely related to modern birds than Archaeopteryx  is!  But there are plenty of other opinions out there if you don't like that one :-)

[Note: sorry if this is all very patronising and obvious to you: I am really writing it for the benefit of other visitors.]

As Mike mentioned, Mayr et al. (2005) produced a phylogeny that showed that birds evolved independently twice. Once as Archaeopteryx and Rahonavis (from Madagascar) and the second as the 'line leading to modern birds' represented by a Cretaceous Chinese bird, Confuciusornis. In their analysis, Archaeopteryx is basal to (Troodontidae + Dromaeosauridae) whereas Confuciusornis was a basal dromaeosaur along with Microraptor. So in this case, modern birds are closer to dromaeosaurs like Velociraptor than to Archaeopteryx. Alan Feduccia and co argue further that dromaeosaurs are actually not dinosaurs at all but secondarily flightless birds.

However, there are some technical debates as to how strongly supported Mayr et al.'s (2005) hypothesis of birds arising twice really is. Corfe and Butler (2006) showed that it isn't any more strongly supported than the alternative, that birds only arose once (Archaeopteryx is closer to confuciusornis than to other theropod dinosaurs).

More recently, Phil Senter presented a phylogeny where Archaeopteryx is closely related to all other Cretaceous birds. In his analysis, Rahonavis is shown to be a basal dromaeosaur.

So I guess you can pick and choose, but I have the impression that the majority of palaeontologists would be happy to place Archaeopteryx as a basal bird, even basal to enantionithines, but related to modern birds.

"Mayr et al. (2005) produced a phylogeny that showed that birds evolved independently twice. Once as Archaeopteryx and Rahonavis (from Madagascar) and the second as the 'line leading to modern birds' represented by a Cretaceous Chinese bird, Confuciusornis."

Well.  That depends on what you mean by "bird".  Mayr's interpretation of his findings is rather idiosyncratic: most systematists would start out with a definition of "birds" (e.g. the crown-group of all living birds plus all descendents of their most recent common ancestor) and then simply ask what does and doesn't fall into the group on the basis of that definition.  So a more mainstream was to interpret May's phylogeny would to say that Archaeopteryx is not a bird.  (I wonder whether that's what the original questioner had in mind?)

Frustratingly, the matter of which definition to use for the group Aves (birds) has been the subject of more argument that probably any other group -- Gauthier and de Queiroz (2001) wrote an entire paper, 35 pages long, on the subject.  One of the candidate definitions -- arguably the one most consonant with historic usage -- would mean that Archaeopteryx is a bird by definition: Aves can be defined as all descendents of the most recent common ancestor of Archaeopteryx and modern birds.  If you use that definition, then Mayr's findings yield the rather delightful interpretation that Velociraptor and Troodon are birds.