What the hell is going on with dromaeosaur tails?


One of the few things that everybody knows about dromaeosaurs - the sickle-clawed maniraptoran theropods best represented by Velociraptor from Mongolia and Deinonychus from Montana - is that they possessed a peculiar tail. Super-long zygapophyses and chevrons formed a bizarre, inter-twined array of body rods that ran the length of the tail and apparently assisted in its function as a dynamic stabiliser [image, © Greg Paul, shows Velociraptor versus two troodontids].

In describing this remarkable conformation in Deinonychus, Ostrom (1969) suggested that these bundles of bony rods would have inhibited flexibility: he imagined that the tail was stiffened and functioned 'as a single unified member' (p. 79). Ever since, people have been depicting dromaeosaurs as stiff-tailed, at most capable of bending the entire tail laterally or dorsally as a unit, but not with any of the degree of flexibility present in other long-tailed theropods.


In view of all this it was very surprising when Norell & Makovicky (1999) described a new Velociraptor specimen in which the near-complete, articulated tail was preserved in a gentle sinuous S-curve (with the curves being in the lateral plane, not the vertical) [tail of specimen shown here, from Norell & Makovicky 1999]. They noted that 'this conformation of the tail seems not to disturb the arrangement of the elongate prezygapophyses, suggesting that there may have been substantial lateral mobility of the tail in life' (Norell & Makovicky 1999, p. 24). It's true that, in the specimen they described (IGM 100/986), the curving of the tail does look natural and 'life-like', but if this really was possible in life it raises the question as to why these animals had the super-long zygapophyses and chevrons in the first place. Elongate, cylindrical bones (like these zygapophyses and chevrons, and like the cervical ribs on Tanystropheus and brachiosaurs for example) must have had some flexibility, but were they really this flexible?

Judging from the recent description of the Mongolian dromaeosaurid Tsaagan mangas Norell et al., 2006 from the Djadokhta Formation, Mark Norell and colleagues are indeed still thinking that the dromaeosaur tail had a reasonable degree of lateral flexibility. While the tail of Tsaagan is unknown, a life restoration in the descriptive paper (produced by Nicholas Frankfurt and reproduced below) shows Tsaagan bending its tail laterally in a manner that does not seem too constrained by those super-long zygapophyses and chevrons. Incidentally, you might not be familiar with Tsaagan, but you've probably seen pictures of it, as photos of the excellently preserved holotype specimen have been used in countless magazine articles about Mongolian dinosaurs (typically labelled as Velociraptor).


Could these rod-like bones really be bent, in life, as much as IGM 100/986 indicates? I find this really hard to believe, but as usual my incredulity is nothing more than an opinion and shouldn't count for much. I can imagine that a long, rod-like bone with a sub-circular cross-section has as much play in it as does, say, a long, slim rod of flexible wood (like a fishing rod), but bending it into an S-shape seems like a real stretch. The problem is compounded by the fact that the in-life flexibility of these structures is so difficult to test: the fossils are no use, and there just aren't living animals with rod-like bones at all like dromaeosaur zygapophyses and chevrons.

Could the twisty shape of IGM 100/986 be the result of post-mortem distortion? We know from many, many fossil bones that structures that were straight in life can become bent and distorted into wiggly or curved shapes when subjected to hundreds of thousands of years of compression. A good example is provided by the neural spines of Spinosaurus aegyptiacus which, when viewed anteriorly or posteriorly, could be seen to be wiggly (Stromer 1915) and not vertically erect as they must have been in life. In the case of Spinosaurus, this happened, presumably, because the tall neural spines were lying on top of ribs or other structures, and were slowly forced to take on dips and wiggles as they were compressed by sediment. But, having said all that, the tail of IGM 100/986 really doesn't look distorted and in fact looks in pretty good shape. Plus, the rod-like parts of the zygapophyses and chevrons of these animals are only 2 mm wide or less, perhaps making flexibility plausible.

So, I'm confused. I don't have any answers and would be interested in seeing what people think. Can long, rod-like bones really be this flexible? And, if so, what's the point of having them in the first place? And should we all stop drawing our dromaeosaurs with stiff, rod-like tails?

Refs - -

Norell, M. A., Clark, J. M., Turner, A. H., Makovicky, P. J., Barsbold, R. & Rowe, T. 2006. A new dromaeosaurid theropod from Ukhaa Tolgod (Ömnögov, Mongolia). American Museum Novitates 3545, 1-51.

- . & Makovicky, P. J. 1999. Important features of the dromaeosaurid skeleton II: Information from newly collected specimens of Velociraptor mongoliensis. American Museum Novitates 3282, 1-27.

Ostrom, J. H. 1969. Osteology of Deinonychus antirrhopus, an unusual theropod from the Lower Cretaceous of Montana. Bulletin of the Peabody Museum of Natural History 30, 1-165.

Stromer, E. 1915. Ergebnisse der Forschungreisen Prof. E. Stromers in den Wüsten Ägyptens. II. Wirbeltier-Reste der Baharîje-Stufe (unterstes Cenoman). 3. Das Original des Theropoden Spinosaurus aegyptiacus nov. gen., nov. spec. Abhandlungen der Königlich Bayerischen Akademie der Wissenschaften Mathematisch-physikalische Klasse 28 (3), 1-32.

More like this

Once more, I'm going to start recycling some of those dinosaur texts written for the defunct field guide (for the back-story on that project see the ornithomimosaur article here). This time round, I'll get through some of the ceratopsians [adjacent skull reconstruction from wikipedia, and based on…
A little while ago, news of a new paper by Devon Quick and John Ruben, both of Oregon State University, appeared on the newswires. It got its fair share of publicity. Entitled 'Cardio-pulmonary anatomy in theropod dinosaurs: implications from extant archosaurs', the paper (Quick & Ruben 2009…
After the success of The Lost World: Jurassic Park, the minds behind the franchise were in a bit of a fix. Tyrannosaurus and "Velociraptor"* had the run of the fictional islands for the past two movies; something new was needed to up the ante. The decision was made to make Spinosaurus the new…
A story of cheeks, beaks, feathers, bizarre theropod dinosaurs, and truly, truly amazing fossils.... Yesterday I made a special visit to the University of Portsmouth's School of Earth and Environmental Sciences in order to attend a talk by, and meet, Professor Altangerel Perle, the famous…

Darren, another possible explanation is sedimentologic diagenesis, but this requires finding unusual bends in long bones or ribs that conform to sedimentologic curvatures.

It is entirely possible that bending occured through postmortem effects, since the tails of dromaeosaurids in North America that are preserved are even shown in vertical arcs, but not sinuously, and the rods tend to be split. Vertical limits on curvature are noted in GI 100/25, the Fighting Dinosaurs specimen, and in tails attributed to Deinonychus from Ostrom's original site and from the Oklahoma associations with a juvenile Tenontosaurus.

The curvatures seem to tend to the vertical, rather than the lateral, so the lateral sinuosity is unusual, exceptional, and likely diagenetic. That the rods are composed of ossified tendons was proposed by Ostrom and upheld through sectioning; they are not normal bone and this affects their potential diagenesis, unlike bone.

By Jaime A. Headden (not verified) on 26 Aug 2008 #permalink

I think thin rods at these sorts of diameters were quite flexible. The thin fibula of my cooked chicken seems to be quite bendy (but is that because it is cooked?) Coelophysoid neck ribs were extended in length and formed overlapping bundles somewhat like the bundles in Dromaeosaur tails, yet numerous articulated coelophysoids are preserved with their necks in all sorts of positions. My pet hypothesis is that bone at that thin diameter was flexible and springy. Put enough together as a bundle and they may have acted as a spring (a bit like the leaf springs of a car's suspension). I could imagine Coelophysis walking around with its neck drawn back by its axial muscles with its cervical rib bundles drawn into a sigmoid curve. When a quick predatory strike was needed it could relax its muscles a little and the cervical bundles would straighten, snapping the head forward. 'Corse this is all just rampant speculation, anyone out there feel like testing it?
A big difference between necks and dromaeosaur tails is that necks are heavily muscled where as the rod like part of a dromaeosaur tail was probably unmuscled and therefore shouldn't have been able to bend under its own steam in the way that reconstruction of Tsaagan shows. So i see their tails as stiff and rod like but were somewhat 'springy' in order to avoid snapping during rough physical activity.

By Adam Yates (not verified) on 26 Aug 2008 #permalink

the tails may have been inflexible, but not straight! Particularly at the tip, which appears to be the case in the photo.

Hwang et al. (2002: New specimens of Microraptor zhaoianus (Theropoda, Dromaeosauridae) from northeastern China. American Museum novitates ; no. 3381) note that the tail of Microraptor is less flexible than those of both Deinonychus and Velociraptor. This is a size-related effect but probably it may be influenced by the more proximal position of the transition point present in Microraptor (= influenced by the reduced number of proximal caudals having developed muscular processes and lacking elongate rods).
As noted elsewhere (by Paul? I don't remember), the dromaeosaurid tail is similar to those of non-pterodactyloid pterosaurs: did microraptors used their tails in a manner similar to "rhamphorhynchoids"? Given that flying-flapping taxa as Microraptoria, Rahonavis and Aves are successive sister-taxa of the "Velociraptor-Deinonychus clade"*, it is possible that the reduced stiffening of Velociraptor's tail was a reversion to a life style more terrestrial (less "rhamphorhynchoid")than those of basal dromaeosaurids.

And what can Jeholornis tell us about? A more detailed phylogeny of paravians is needed to map the distributions of these features...

*: Ok, there are "non-volant forms" placed along the lineage leading to "true dromaeosaurids" (troodontids, big-unenlaginae...), so the evolution/distribution of flight-flapping abilities in Paraves is a more complex phenomenon...

Prehensile! Picture them swinging through trees, and dropping onto their prey.

So, when do I go pick up my Noble?

Well, just adding my 2 cents, I think we would at least have to find more than one in this condition. Also, what were the conditions it was buried in? Maybe the tail was quickly buried in sand or heavy earth and bent on death and stayed that way.

The area that seems to have the most tendons and where the curve is looks "uncomfortable" to me. It's a little thicker in that area almost as if some of the long thin bones may be broken.

Well, I have zero paleontological or zoological credentials, so this is a very uneducated opinion!

On the subject of dromaeosaurs, but not their tails, what is all this about there strange way of breathing - breathing in but not breathing out, witht the air instead "exiting the body through holes in some of the bones" as quoted from 'Dinosaur George' Blasing in 'Jurassic fight club' (talk about a dumbed down title!)

I'm unsure if stiff tail is adapted for balancing, given that cats and monkeys balance with tails and have it flexible.

Just some questions:

- could tail support the body when dromaeosaurid climbed up tree trunk?
- could tail have two rows of tail-feathers and be stiffened to provide gliding plane?
- could dromeosaurids be momentarily supported by tail when kicking forward with Famous Slashing Claws?

BTW, Velociraptor has curious adaptation to point tail skywards. Becomes obvious when you forget Jurassic Park and see Meerkat Manor.

Actually, since tree climbing has bee brought up, could dromaeosaurs climb trees? I know the ripping and slashing sounds great, but could they have used the "sickle claw" in tree climbing? I always wondered if it could be used the way the telephone pole guys used to wear special shoes with spikes to climb poles many years ago.

I know. I'm veering off topic...

I'm all for the tree-climbing. I think smaller, arboreal microraptorines may have used their tails for two purposes: flight (as in rhamphorhynchoids) and providing a resting point while on the trunk. Perhaps microraptorines had large, thick feathers at the ends of the tail, so they could "lean back" if you will, much like a woodpecker does.

The tails of basal dromaeosaurs may have become stiffened for this purpose, and later dromaeosaurs simply inherited the condition. I doubt the S-curved tail up there is how the tail would behave in life. I'll bet that dromaeosaurs used their tails for balance while running and turning quickly (as a cheetah would) and, perhaps, display.

Very thin bone is flexible in any case. Think of prokinesis in birds: the upper beak can move upwards independently of the rest of the skull. Except in parrots there is no joint for that. Instead, the solid, plate-shaped bone bends. I've done this myself on an apparently relatively fresh flamingo skull. Over time, bone dries out and becomes hard; it didn't work on any other bird skulls in the collection.

By David Marjanović (not verified) on 26 Aug 2008 #permalink

I would think that the flexibility depends partly on the density and orientation of collagen fibers. Would be pretty easy to determine with bone histology of these rods.

My very limited knowledge of bone material properties suggest that it can be quite flexible, so I expect there was quite alot of play in dromaeosaur tails.

jck, that was my thought.

How much muscle was there on the dromaeosaur tail anyhow? If the tail was flexible, but really really springy due to all of those ossified tendons, it might have been that they couldn't have bent it much if they didn't have the muscles to produce enough force to pull it out of shape.

That is a very weird specimen of V-raptor.

I have always wondered about the tail too. No useful input.

However a different question I've been meaning to ask people. I encountered some feed back about my 3D raptors from someone claiming that Dromaeosaurs couldn't move their legs more forward than when they were straight below themselves. As in the legs could only move backwards

This makes next to no sense to me.

The only thing I'd heard (from Phil Currie) is that their legs bent inward so that the feet touched down almost below their mid line. He thought this showed some evidence of them hopping like kangaroos

I guess Raptors (their legs in particular) have me confused. Especially given those tracks just found in Korea look like normal theropod tracks (apart for two toed) to me

Any insights anyone?

Darren - Chris Organ did a finite element model of ornithopod tails and found that they were surprisingly flexible. Ornithopod tails were completely capable of making _Alligator_ like mediolateral undulations without interference from the ossified tendons. It might be possible that the elongated chevrons and zygapophyses of dromaeosaurs functioned similarly in that they may have been used mostly to reduce dorsoventral movement, rather than all directions.

That we are now seeing evidence of greater tail mobility in dromaeosaurs does not come as too much of a shock. I think this specimen gives us an idea of their potential flexibility, but as Bob B. wrote, it should still be viewed as unique until we find more specimens with similar degrees of flexibility.


Organ, C. 2006. Biomechanics of Ossified Tendons in Ornithopods. Paleobiology, 32(4):652-665.

Has anyone done a comparison with the tails of kangaroos? In them, the tail I believe acts as a store of energy, making the bounding gait highly energy efficient. If you watch birds move around on the ground, many hop rather than walk - perhaps some raptors did the same? - at least at speed


To start off with cheers. There's a fun fact I'm adding to the list.

Though this person was very firm on Dromaeosaurs not being able to move the leg forward... Which in light of your info makes their contention all the more confusing!

Is that JUST large theropods who couldn't move them too far back or was it specific family groups (eg. coelurosaurs)

Well, this is cool news, but over 5 years ago a few of us on the dinosaur mailing list realized that the dromaeosaur tail could have been very flexible. At the very least, not a bone-stick proto-pygostyle. http://dml.cmnh.org/2003Feb/msg00202.html

That said, I wonder if the reason we see this sigmoidal curve in this specimen is due to it dying in a burrow (of its own, or another creature). If that's the case, the sediment should resemble what was seen for Oryctodromeus.

@Traumador: Moving femurs forward is equivalent to leaning the head to the ground. Moving femurs back is equivalent to standing erect. In a contest between your informant and Zach, Zach wins on the merits.

By Nathan Myers (not verified) on 27 Aug 2008 #permalink

Traumador, I'll have to go through my archive, but the paper I have about femur movement in theropods specifically states that maniraptors had a fairly avian leg movement arc. That is, more motion in front of the vertical than behind.

Jura wrote:

Chris Organ did a finite element model of ornithopod tails and found that they were surprisingly flexible. Ornithopod tails were completely capable of making _Alligator_ like mediolateral undulations without interference from the ossified tendons. It might be possible that the elongated chevrons and zygapophyses of dromaeosaurs functioned similarly in that they may have been used mostly to reduce dorsoventral movement, rather than all directions.

I think Organ's thesis here is tempered by the fact that ornithischian tails are bounded by a system of latticed tendons, which is inherently more flexible as a system than are those of a series of long rods which are perpendicular to the ranges of movement and constrained into bundles. The flexibility of the tail has a major constraint on movement due to the zygapophyseal facets and the ligaments that bound these to preceding and succeeding elements, not just the rods; in ornithischians, you have a higher grade of movement just because of the vertebrae, with a limiting although flexible system surrounded them. Not so in dromaeosaurs, which have systems not unlike that in sauropod necks (esp. omei/mamenchisaurs) but doubled above AND below the centra, and to either side. Flexibility of the tendons or processes extending from the bony corpus is simply another limiting factor.

By Jaime A. Headden (not verified) on 28 Aug 2008 #permalink

For me ALL the dinosaurs could bend their tails,more or less...Don't forget that they were animals,and i've yet to find a so called "rigid" animal ;)

Plus,with tails like that,it's ludicrous to think that they couldn't move 'em at least to some degree.

Editorial nitpick here: Don't you mean "conformation" in the sentence below?

"They noted that 'this confirmation of the tail seems not to disturb the arrangement of the elongate prezygapophyses, "

Is that JUST large theropods who couldn't move them too far back or was it specific family groups (eg. coelurosaurs)

No. It's all dinosaurs without exception to this day, as far as I know.

By David Marjanović (not verified) on 31 Aug 2008 #permalink

It also looks as if there is a degree of twisting in the distal end of the tail, or am I reading that wrong? How much twist could dinosaur tails have? Of course that varies greatly per species, but lets just use theropods, avian and non as an example.
Working with animators a lot, I often describe flexibility in the tail of dromaeosaurs to be similar to that of a fishing rod. Is that a good analogy?

On the subject of dromaeosaurs, but not their tails, what is all this about there strange way of breathing - breathing in but not breathing out, witht the air instead "exiting the body through holes in some of the bones" as quoted from 'Dinosaur George' Blasing

OMG. Seriously. Did he really say that? Because if so, that's about the most radical possible misunderstanding of pneumatic bones and bird-like breathing. And to think this is on a supposedly educational program in 2008. Shudder!

This might be too long and boring a comment on a fairly old post, but what the heck...
I've grown up with stiff-tailed dromaeosaurs, and, because of those extra-long zygapophyses and chevrons, the idea always seemed to make sense. However, I must admit that the tail of IGM 100/986 looks very life-like. I mean, everything is articulated and nothing is broken, so... Yes, I think that we should stop showing dromaeosaurs with rodlike, inflexible tails. The real question for me is, WHY did they have those pesky zygapophyses and chevrons if they did'nt even do much to stiffen the tail? The world may never know... Anywho, I have always thought that we were underestimating the flexibility of many dinosaurs' tails. After all, how is the tail of a tyrannosaur SO much different from a Komodo Dragon's that that the tail couldn't have been similarly flexible? As Alessio said, "rigid" animals are non-existent.

By Michael Erickson (not verified) on 31 May 2009 #permalink