Thanks for your answers (I think!) Part of my original question did indeed hinge on the idea that Archaeopteryx may not have been a bird at all. Calling it one 'by definition' seems to be a circular argument to me, and pretty artificial. (It also presents the problem of having to change the definition every time a more basal 'first bird' is discovered.) Considering that it has far more 'dinosaurian' characteristics than 'avian' ones, (non-fused vertebrae, no keratinised bill, no metacarpal fusion, cervical vertebrae meet the skull at the back, not underneath, and so on, including feathers, apparently a dinosaur feature, not just an avian one, and so therefore neutral in this discussion) why is it considered a bird at all? Is flight (however weak) seen as the hallmark? Bats fly, but this doesn't remove them from Mammalia, and they are totally unlike any other mammals skeletally.
Also, if basal to troodontids and dromaeosaurs, some cladograms would therefore put it as basal to almost all maniraptorans, inc. therizinosaurs. I've even seen virtually all coelurosaurs included as maniraptorans, so tyrannosaurs would also be in there. Dilong seems an unlikely flightless bird, but I must admit Sinornithosaurus appears more birdy than any terrestrial animal has a right to be!
So, (sorry for all this meandering!) I guess the question boils down to why is Archaeopteryx considered a bird - historic convention, cladistic convenience, or solid 'synapomorphic evidence' (I think I know what I mean by that.) I guess its a question that sets a cat among the dinosaurs!

Andrew Plant wrote:

Also, if basal to troodontids and dromaeosaurs, some cladograms would therefore put it as basal to almost all maniraptorans, inc. therizinosaurs. I've even seen virtually all coelurosaurs included as maniraptorans, so tyrannosaurs would also be in there. Dilong seems an unlikely flightless bird, but I must admit Sinornithosaurus appears more birdy than any terrestrial animal has a right to be!

Just to address a particular point: As you mentioned, Maniraptora includes other theropods like therizinosaurs, oviraptorosaurs etc. and can be controversial some times, but I'm pretty sure that tyrannosaurs and other basal coelurosaurs are not included in Maniraptora.

I think the general concensus now is that troodontids and dromaeosaurids form a clade of its own known as the Deinonychosauria, which Archaeopteryx and other birds are basal to. Birds and deinonychosaurs together generally form a clade called Paraves, which therizinosaurs and oviraptorosaurs are basal to (there are conflicts in the exact positions of these two in relation to Paraves). And this goes on till we have the clade Maniraptora, Maniraptoriformes, Coelurosauria, etc....

Hi, Andrew.  Let me try to address some of your issues.  These problems come up whenever anyone asks "is X a Y?" but they are particularly problematic when Y is "birds" because the group of birds -- Aves to give it its scientific name -- has had so many definitions.  If you pop over to the Phylonames message board, at
you'll find a poll on which is the best of several candidate definitions of Aves ... there are six options on show, and there have been 98 comments on the poll!

When dealing with the question "Is X a Y?" there are two basic ways to approach the problem.  One is to just to use your general sense of whether X is Yish: for example, traditionally as lot of people have said "Archaeopteryx has feathers, so it must be a bird".  This works OK in any individual case, but tends to lead to nasty messes when applied to lots of species; it also means that the answer to the question is very much dependent on personal preferences.  For example, I might take the position feathers => bird, while you might argue that a keratinised bill is a more birdy feature, so that Archaeopteryx's lack of one means it's not a bird.  We can go back and forth on that all day (or for our whole careers :-)

So the second approach is unsurprisingly gaining ground in vertebrate palaeontology and some other fields of biology: that is to begin by agreeing a rigorous evolutionary definition of Y, and then evaluate the evidence of whether or not X satisfies that definition.  In this case, we define Aves, then use objective critera to decide whether Archaeopteryx is a member.  This will certainly not put and end to disagreement (as if!) but it fundamentally changes the nature of the dispute.  Instead of arguing whether feathers or bills are more the birdy character (a matter for philosophers rather than scientists), we will argue about whether Archaeopteryx is more closely related to modern birds than Velociraptor is -- a question that can be addressed using evidence.

So here's how we do it in Phylogenetic Nomenclature, which is the name of that system that's growing in popularity.  We start with a taxon name such as Aves or Dinosauria or Flagellicaudata, and we tie it to a clade: that is, to a chunk of the tree of life, consisting of a single ancestor plus all its descendants.  How to nominate that ancestor is an issue in itself: there are at least half a dozen methods, but let's just pick one: a node-based definition is of the form "A and B and their most recent common ancestor plus all its descendants".  (There are other wordings but they amount to the same thing.)  For example, Dinosauria might be defined as "Iguagnodon and Megalosaurus and their most recent common ancestor plus all its descendants".  What this means is that you trace up the tree from those two dinosaurs and keep going until the two lines of ancestry meet.  That meeting point is the most recent common ancestor, and ALL of its descendant are BY DEFINITION members of Dinosauria.  So we then don't need to answer questions like "Should Eoraptor be considered a dinosaur"; instead we can ask the less fluffy question "Is Eoraptor a dinosaur?", meaning "is it a descenant of the most recent common ancestor of Iguanodon and Megalosaurus?"  That is a question that can be addressed by science.

So now we come back to your original question, is Archaeopteryx a bird?  In order to answer that scientifically, we need a definition of "bird" (that is, of the clade "Aves").  This of course is where it all falls down at the moment, because as I said earlier, the definition to use for this particular clade happens to be hotly contested.  This will get resolved in a year or two by the publication of a book called "Phylonyms: a Companion to the PhyloCode", which will contain definitions for many important clades.  These definitions will be the ones recognised by the PhyloCode, the formal code of nomenclature that will govern phylogeneric names.  As of this writing, it seems likely that the definition to be canonised will be that of the crown-group: that is, all descendants of the most recent common ancestor of all currently living birds.

Finally, we answer your question!  According to all published phylogenies, Archaeopteryx lies outside that crown group, so according to what is most likely going to be established as the phylogenetic definition of Aves, Archaeopteryx will not be a bird.

I'm sorry if that was a long way round -- I hope the ride was fun.  Maybe we should have have answered "No."  :-)

As a postscript, I'd like to address your question "Calling [Archaeopteryx] one 'by definition' seems to be a circular argument to me, and pretty artificial".  It certainly seems artificial is the question you're asking happens to be "is Archaeopteryx a bird?" but a better word might be "arbitrary".  That definition is a perfectly good one, and leaves open the question of whether Velociraptor is a bird.  (Answer: if and only if it is found to be more closely related to modern birds than to Velociraptor.)  But we would never "change the definition every time a more basal 'first bird' is discovered" because the whole point of phylogenetic nomenclature is to nail down clades, associate names with them, and leave them there.  So when something more basal than Archaeopteryx is discovered, it would not be considered a bird under the Archaeopteryx-anchored definition, no matter how "birdy" it was.

OK, that's it.  I really have finished now.

Hello all,

I have been trying to think how best to respond to the growing interest in the 'is Archaeopteryx a bird?' debate, and I think that the best way that I can add to this line is to talk about the importance of the fossil first discovered in 1860, represented by a single feather.  Archaeopteryx is by far my favourite fossil.  I don't work on birds, or dinosaurs, but nevertheless the small number of specimens, (now about 10) skeletons and the single feather (now in the Humbolt Museum für Naturkunde in Berlin) could not have been discovered at a more important time for the understanding of evolutionary biology!!

in 1859 "The Origin of Species" was published to rapturous applause, or shock and disgust.  The idea that all life evolved from earlier forms was pleasing and disturbing to different groups, in equal measure.  Before Dawkin's books, or such like, were doing it, Darwin set the precedent, by selling out the first print round, and going straight to the top of 'must have books', or it would have done so if the Victorians had had such lists.  Needless to say it sold out immediately.

Origin describes how lifeforms evolve from one to another, depending on the environmental pressures applied to the population.  However, many asked where the evidence was?

The discovery of an almost complete specimen in 1861, preserved in the lithographic limestone (very fine limestone used for producing printing material) in Solnhofen, Germany (the same area as the feather  had been found two years earlier) was what was needed to help support the claims made so well by Darwin. Archaeopteryx takes its species name from the limestone it was found in, Archaeopteryx lithographica.

The London Specimen, as it became known, lacks most of the head and neck, but crucially it has a reptilian body, covered in feathers.  In 1861 feathers were a 'bird only' character, that is to say, if you have feathers you are a bird. 
The discovery of what is known as the Berlin specimen in the 1870's is the specimen that people 'see' when they think of 'the' fossil of Archaeopteryx.  It is complete, with all of the feathers, and the skeleton, remarkably intact. 
So, I can add little to the overall debate on whether or not it is a bird (I think it is, the feathers are well formed, and it could probably fly better than me).  Sure it lacks a beak, and the wishbone is not very well developed, and I am sure that you can think of dozens more reasons why when it comes to claiming to be a bird, it's a bit of a 'turkey' (that is a poor joke, for which I apologise), but in the case for evolution, its preservational quality and timing of discovery could not have been more important.  If the feathers had not been preserved it would have looked like any other small theropod dinosaur (as some specimens have been wrongly classified), but with the feathers its a 'transitional form' and makes a wonderful case for evolution.   

If you want to read more then Wikipedia has a page or two worth looking at (link below) 

and 'Taking Wing', by Pat Shipman is a very good book to read.

Hi Neil. I agree about the importance of Archaeopteryx as an illustration of an evolutionary transition back in the C19th, but science has moved on a lot since then. Small Theropod dinosaurs had feathers (and some had wishbones), so the only Avian "character" of Archaeopteryx is that its feathers were asymmetric (thus suitable for producing lift). Microraptor also had asymmetric feathers (although it is not considered a "bird") and it is even better than a "transitional form" because it gives evidence of a potential pathway for the evolution of flight.

Perhaps we should just stop referring to Aves and deal with birds simply as members of the Therapoda when working at a higher taxonomic level, rather than shoe-horning the evidence into preconceived structures of understanding. We don't have these problems with the bats (we have a different set of problems with them!).

Last edited by Paolo Viscardi (14th Jun 2007 08:42:02)

Hi Paolo,

I agree with what you say entirely, and would have said as much, had I not run into a few problems with the upload of my answer to this post, that meant that I have only now been able to review it.

Archaeopteryx, as far as I can see, was named for the feather found in 1860.  The feathered animal called Archaeopteryx was named because it was felt (at the time) that the feather came from it (although this is now some limited debate about this), therefore, because birds have feathers and Archaeopteryx has  feathers, Archaeopteryx is a bird, and in the 19th century the first bird!  I agree that there are numerous theropod dinosaurs with birdy-wishbones, and wrists, and even feathers that are not birds, but to the Victorians, and evolutionary biology the discovery was both timely and spectacular.

Obviously now we have all of the earlier forms that are more bird like than Archaeopteryx, and as you say birds are really avian theropod dinosaurs.

I hope that my previous post simply supplied some information about the history of Archaeopteryx, and a great book to read on the subject.

Maybe we can call Archaeopteryx the 'First (Dino)bird (to be discovered)'.

Last edited by Neil Gostling (14th Jun 2007 09:32:07)

By the way, this is one reason why the crown-group definition of Aves may be preferable: it makes it very clear that Archaeopteryx, Microraptor et al. are NOT birds, and saves us a lot of harmless but time-wasting semantic quibbling on the point.

Ah, but Mike where would science (and *especially* taxonomy) be without semantic quibbling?

Lets face it, few people can agree on the definitions of some very common and well studied animals (are they species, sub species, races etc.) and species definitions vary between groups anyway - a bacterial species is not equal to a dinosaurian or mammalia or botanical one.

More seriously, much time is often wasted on these arguments (both science and non) and my opinion would be that just sticking a name on a (stable) clade and leaving it should be fine. As long as everyone agrees on the definition it should not matter what is in and what is out of the clade. I think the problem lies with Archao and the birds - they are such an evocative group (the Berlin Archaeopteryx is probably the single most identifiable fossil in the world and one that even young kids can see has feathers but also a tail and teeth) and so important for the fundamentals of taxonomy and evolution that I think the problem has moved away from science and into the realms of semantics as a result.

I'd happily start bids with Archaeopteryx as a rather romantic statement using its historical and scientific importance as a fossil to define the clade. If that means Velociraptor is a flightless bird, who cares? So are orstriches and penguins. I don't think there is anything in the ICZN that would prevent it and if we have to draw the line somewhere, why not use something that any scientist (and the public that matter) are happy with as a frame of reference, and not just the theropod nerds?