Anchiornis. Modified from Hu et al., 2009.
OK by now the ‘secret’ is probably out and the ‘new basal avialian’ that my colleagues and I described earlier this year, Anchiornis, turns out not to be an avialian (or avian if you prefer) at all, but in fact is a very basal troodontid. Those new specimens I mentioned at the time brought in a ton of extra detail and information that allowed researchers to firm up the diagnosis of this animal and show conclusively that it is indeed a troodontid. There are some important and interesting lessons to take from this, both in terms of theropod and bird relationships and how new information changes perspectives. To try and avoid me rambling on and keeping the issues clear, I have listed them:
1. Our initial assessment was not ‘wrong’ as such, we merely did not have all the information that would have revealed the troodontid identity of this animal. The original diagnosis was well supported by the information we had and the accompanying systematic analysis. We rarely have complete animals to work from and often have to work out an organisms ancestry based on a few key characteristics, if those are missing it can be very tricky.
2. In this case the missing piece was the skull which, for archosaurs, is incredibly important in learning about an animal. For cladistic analyses at least half the characters often reside in the skull, jaws and teeth, so with that missing you immediately have your information cut in half. Add to that any other missing pieces, distortion or compression of the fossil, pieces hidden under each other, or split and broken, or of course only having the thing in 2D because it’s preserved in a slab of rock and not as a 3D set of bones and you can see why we can’t always make as accurate designations as we would like.
3. By extension of the above points, this shows just how similar taxa are at the point where dromaeosaurs, troodontids and avialians (which together make up the Paraves) diverge. There are still longstanding discussions and conflicting analyses over exactly which groups is related to which and in what way. Even with detailed analyses using complete who skeletons some of these relationships are still subject to debate. If anything Anchiornis clouds the issue a little and certainly illustrates how intimately linked these groups are.
4. While we already knew that troodontids had feathers, including on their legs, this is a particularly good example and does extend our knowledge of how these evolved and were distributed around the theropod tree. Since we now have basal avialians (Archaeopteryx), basal troodontids (Anchiornis) and dromaeosaurs (Microraptor) with feathered legs then it seems clear that this is likely to be a primitive characteristic for paravians and that many, and possibly all, of them had these kinds of plumes.
5. However, it does not mean that they were all parachuting, gliding, or even flying – merely that they had long leg feathers. The feathers on Anchiornis are symmetrical and thus provided no (or at least low / minimal) lift and the animal shows none of the specialisations of Microraptor to gliding or of Archaeopteryx to powered flight. It does raise the possibility that other paravians developed something similar however given the fact that they had the apparent exaptations (also called pre-adaptations) of elongate and broad leg feathers.
The paper itself focuses more on the age of the fossil and the implications there of early troodontids and I’ll be returning to this shortly in another post, but right now I’m at SVP and thus rather busy.
Share this Post
Dave Hone's Archosaur Musings 
I walked into the Wills as this slide was being switched, so my mind was blown, but I was so angry that I missed it! Thanks for posting the picture, sir, and it was great to meet you!
Well there are better images than this in the paper and as with *last* time, there are more specimens out there again already, so expect more in the future.
Are there any craniates for which this isn’t true?
A good point! I overstated a little with a bit of a tautology. Though some are certainly more important than others – to turn (inevitably) to pterosaurs, derived pterodactyloids often don’t have much going for them in the skull (relatively) and a good post-cranial skeleton can be diagnostic – something not (necessarily) true with dromaeosaurs and birds for example.
hallo, I am new here, but I just want to say, I am a big fan of your website, the informations here, are really amazing. I want to ask, if you can confirm the callovian age of Anchiornis?
I’m not sure what you mean by ‘confirm’. I trust my colleagues to get the dating right if that’s what you mean. I freely admit that I am no geologist and although this has been a contentious issue for a while, my understanding is that these dates are now pretty solid and based on good dating of strata above and below the fossil deposits, but I really can’t add anything that isn’t in the paper.
ok, I understand, I just heard some rumours about the age, and I was just wondering, because if it´s true, the ´temporal paradox´ problem would be solved. I´m very interested in this, because I am now writing a book on the origin of birds and I need every relevant information for the book. thank you.
Lovely preserved. What kind of stone is this? marble? or quartz!?
Neither, they don’t tend to preserve these kinds of tissues. I actually don’t know the exact composition since I’m far from a geologist, but the Dahougou beds are primarily made from volcanic ash which cretainly does preserve both bones and soft tissues exceptionally well.
You might be amused (or not) to know that a creationist is accusing you (or Hu et al) of “drawing” on the feathers in this fossil:
http://www.talkrational.org/showthread.php?p=782689#post782689
So I see from the threads linking to this page. It happens. You could point him to my post on feathered dinosaurs that lists all the various specimens, or the UV photos of Archaeopteryx on here that shown in at least some cases that there are real feather traces and not just impressions. I actually have a paper in review on this as well and as it happens I’m off to see Anchiornis again this weekend.
Short version, he has not seen the specimens. I have seen 7 speciemns of Archeiopterux, 8 of feathered Caudipteryx and Protarchaeopteryx, 3 of Anchiornis, a dozen Microraptors, Dilong, Epidexipteryx, several Sinosauropteryx, and dozens if not hundreds of feathered birds. I’ve put these under microscopes and UV light. I’ve exmained and sampled them and so have dozens of other researchers. No-one can paint or scuplt the details that are present, and not in the vast numbers that these fossils come out in (there lots more undescribed or privately held than are in scientific papers) and I know people who’ve found stuff themselves and dug them out with their own hands.
The stuff is not fake and anyone assuming they are has simply not the knowledge and understanding of science to know better or lacks the confidence to actually go and see for themselves. Many of these are on display in China, Europe and the US so he can toddle off to a museum and check. But then he’ll have to make himself either a liar or an idiot to maintain that they’re fakes.
And while I’m thinking about it, while I *really* don’t want a creationist to take up residence on these pages, I’d be intrigued to know if he thinks that just the feathers are fakes or all the soft tissue stuff from the Solnhofen and Liaoning – the tens of thousands of fish, amphibians, insects and things like pterosaur wings, furry mammals and so on. It that all fake too, or are only the feathers on dinosaurs fake but not the feathers on birds?
Looking over his comments, he looks like a kook and / or idiot. His mind will not be changed no matter how much evidence and explanation you give him. His knowledge of the bird-dinosaur transition is terrible / biased / out of date and I don’t think the efforts you or others are making will help much, sadly, though of course it might inform others.
Thanks Dave. I found your articles on UV photos of Archaeopteryx but the image links seem broken at the moment?
are there any with intact photos?
Sorry they seem to be up again now – maybe a server glitch?
Could be. In the meantime, note the new paper out today on dinosaur colour. Again, looks like those ‘fake’ feathers have the structure of modern feathers that only show up on SEM level resolution. Damned funny paint that…
Yeah it’s a nice find. As you state though, the guy is impervious. It does seem to be informing others though so that’s a bonus.
Aye keep it up. I was tempted to send you a huge e-mail covering everything but frankly it seemed like a waste of my time. However, since you and others are clearly interested, i would pint you towards various papers by Tischlinger, including one that shows that the Berlin specimen has both imprints and actual feathers preserved, I have a paper accepted by PLOS1 (out soon!) on the feathers of Microraptor, the recent book by Wellnhoffer covers Archaeopteryx in enormous detail and look out for Faux & Padian on opisthotonic postures (how things look when they die) as that has bearing on that stuff too. Oh yes, and while Archaeoraptor was a fake, it was a composite of two genuine fossils. If you glue a chimp skull onto a gorilla body you still have two real skeletons present.
Thanks for that. He did try and use a composite argument when he noticed that there is a large crack in the London specimen. He tried to argue the 2 parts don’t line up but when we pointed out both parts have feathers so even if it was a composite of 2 fossils they would still both be of dinosaur-like creatures with feathers he quickly moved to another subject.
He does rather bounce from subject to subject. And seems to be reliant on the fact that one day everything will suddenly be admitted to be fake. Oh and he really does seem to confuse diagenesis and taphonomy with evolution which is odd to say the least. And seems to think that if you somehow removed Aracheopteryx the the bird dinosaur link collapses. His knowledge of the science and literature is absolutely pitiful frankly.
You will love this. He’s now convinced himself birds evolved from pterosaurs (specifically Pteranodontids). This is high comedy
Oh dear. Seeley was the last to hold onto that and even he kinda gave up in the early 1900s. Um, that is, well, interesting at best is all I can say. Violates dozens of characters at worst. I don’t even know where to start and I’m not sure I want to.
Oh dear, and I’ve spent the last two days looking at Anchiornis specimens too….
OH NO! Was this the evolution fraud?
Love this blog, particularly this post.
Judging by the threads Monad is referring to, it mostly seems to be the fact that someone does not realise that some fossil taxa are represented by more than one specimen. He thus concluded that anything o9ther than the holotype must be a fraud. Which is terrible logic at best and a terrible misunderstanding of the fossil record and basic taxonomy.
“Since we now have basal avialians (Archaeopteryx), basal troodontids (Anchiornis) and dromaeosaurs (Microraptor) with feathered legs then it seems clear that this is likely to be a primitive characteristic for paravians….”
The similarities between these taxa, and the fact that they not only have feathers, which is a unique feature not likely to have evolved twice, but they also have feathers on their legs, suggest that they are all birds, not dromaeosaurs and not troodontids. As Larry Martin and others have pointed out, a dinosaur with feathers sticking out of its legs are not going to run or walk well on the ground. Therefore they are almost certainly arboreal animals. Feathers make sense for arboreal animals because they can save energy even if they only glided. It simply cost less energy to glide than to run down a tree, cross the forest floor and climb up another one. Only those animals who spend a lot of time on the ground, such as fox squirrels, would find gliding unappealing.
When you classify Microraptor as dromaeosaurid and Anchiornis as troodontid, you are simply suggesting that the terrestrial Cretaceous dromaeosaurid and troodontids are simply flightless birds, even if you don’t realize the ramifications of your classification. The althernative of separate lineages of theropods and birds evolving not only feathers but hindlimb wings independently would be most unparsimonious.
Finally, Anchiornis does nothing to eliminate the temporal paradox. If anything, it makes the temporal paradox worse, because now you claidists have to find ground dwelling theropods that are older than Anchiornis, and that are sister taxa to the Anchiornis-Microraptor-Archaeopteryx clade. So far there is no Jurassi theropod that remotely resembles these small, arboreal feathered, 4 winged birds. There is some good news though. Many cladists are now being pulled (screaming and kicking?) towards the inescapable conclusion that flight evolved from the tree down, from small archosaurs that were arboreal. That simply makes those cladograms whih showed such ground dwelling theropods as Velociraptor and Bambiraptor as close relatives of birds all the more silly, if not downright inane.
“The similarities between these taxa, and the fact that they not only have feathers, which is a unique feature not likely to have evolved twice, but they also have feathers on their legs, suggest that they are all birds, not dromaeosaurs and not troodontids.”
But I am not suggesting that they evolved multiple times. Feathers evolved once, in maniraptorans, and then were retained but later descendants. Hence paravians have feathers. They have similar morpholgies becuase they are close relatives, just like humans are like gorillas and chimaps.
“Feathers make sense for arboreal animals because they can save energy even if they only glided. It simply cost less energy to glide than to run down a tree, cross the forest floor and climb up another one. ”
But that assumes that the feathers have already evolved to the point that they can serve a useful function in gliding. Unless assymetric, shafted feathers appeared in a single generation, this point doesn’t work. You have to get the feathers first. And in any case, as noted, the feathers of Anchiornis are symmetrical.
“When you classify Microraptor as dromaeosaurid and Anchiornis as troodontid, you are simply suggesting that the terrestrial Cretaceous dromaeosaurid and troodontids are simply flightless birds, even if you don’t realize the ramifications of your classification.”
No, just no. Go and read up on how phylogenetics / taxonomy works and this is just not true. Humans are humans and chimps are chimps and goriallas are goriallas. They are all apes, yes, but gorallis and not un-humans. They are close relatives, not derived versions of humans.
“Finally, Anchiornis does nothing to eliminate the temporal paradox.”
Yes it does.
“The althernative of separate lineages of theropods and birds evolving not only feathers but hindlimb wings independently would be most unparsimonious.”
But that is NOT being suggested. Feather evolved once, in tetanurans (around the tyrannosaur / compsognathid node) and then were retained by their relatives and descendents. Where does that suggest multiple origins?
“now you claidists have to find ground dwelling theropods that are older than Anchiornis, and that are sister taxa to the Anchiornis-Microraptor-Archaeopteryx clade.”
Well, allowing for the fossil record and assuming they are all arboreal. And to that in any case I say “Haplocheirus”.
“Many cladists are now being pulled (screaming and kicking?) towards the inescapable conclusion that flight evolved from the tree down, from small archosaurs that were arboreal. ”
Then I look forward to seeing a list of those people and their respective papers. “Other people now agree” is not an argument.
“That simply makes those cladograms whih showed such ground dwelling theropods as Velociraptor and Bambiraptor as close relatives of birds all the more silly, if not downright inane.”
You seem to assume some form of Dollo’s law. That once arboreality has evolved it can’t be lost or can’t evolve multiple times. Velociraptor is a derived dromaeosaur and could easily be secondarily terrestrial. Look at rodents or lizards, multiple forms have become arboreal at different times, or secondarily come down from the trees or do both.
Without wishing to be rude, I really don’t think you understand the issues here. Certainly you are grossly misrepresenting the position taken by cladists over the phylogenetic position of birds wrt theropod dinosaurs. You agree of course, welcome to disagree, but don’t waste my time with strawman arguments of non-positions. Go and read the literature properly and you will see that this is simply not what is being argued. If you are looking for an argument, please go elsewhere. If you are looking for information, try the excelent discussions and details on Tetrapod Zoology, Talk Origins and the Evowiki. Wikipedia is actually quite good too.
Hi Dave
it may amuse you to note that the chap we discussed earlier now has a “rival” blog to yours:
http://pterosaurnet.blogspot.com/
and refers to himself (probably without the slightest sense of irony) as Dr Pterosaur
clearly his delusions are incurable (as we surmised)
Ah – I see now that you have already engaged with him here (he is your “Jack”) in case you hadn’t noticed 🙂
See the comments here:
and my post here: https://archosaurmusings.wordpress.com/2009/10/01/anchiornis-and-the-temporal-paradox/#comments
I’m well aware of what he is doing and am very unhappy about it.