Most dragons (before Skyrim and GoT) had six limbs in total, four legs and a pair of wings.

The question is how would a dragon's forelimbs be placed? It's obvious they have good terrestrial capabilities, without having long noodle legs. But then there's the flight muscle, it needs lots of space and a large attachment site, same goes for the wings. But how should I put the front legs on the dragon so that it doesn't interfere with the wings' motion? Sure, those things only move during climb out, but it' still troubling. Whatever I choose should be compact.

How and where would the front limbs of the dragon connect to the rest of the skeleton?

  • $\begingroup$ Depends on the dragon's attitude in flight. The wings must attach to the spinal column so that when the dragon assumes the flight attitude the attachment is above the center of gravity (or maybe a little forward of it if the dragon has a suitable tail which can act as an airfoil). Conversely, if you fix the attachment of the wings to the spinal column you can assume that in flight the dragon will assume the attitude which brings the center of gravity below the wing attachment point. As for the muscles, they don't interfere; flight muscles attach most likely to the sternum, unlike limb muscles. $\endgroup$
    – AlexP
    Apr 6, 2019 at 17:44
  • $\begingroup$ There is no way to answer this question where it would connect would depend entirely on how and where the limb evolved, what joints it has, and where the muscles are ect. flight musculature is not uniform in vertebrates. Birds and bats have a very different layout of flight muscles for instance. Since we don't know these things we can not give a definitive answer, there are many many correct answers. The only thing you can do is lay out the limbs so their range of movements do not meet each other. $\endgroup$
    – John
    Apr 6, 2019 at 21:33
  • $\begingroup$ I can suggest you look at the art from, A Natural History of Dragons: A Memoir by Lady Trent, it has probably the most sound anatomy I have ever seen for dragons, with a minimum of interference between the limbs. This will show you at least one way it can be done. $\endgroup$
    – John
    Apr 6, 2019 at 21:38
  • $\begingroup$ But do dragons really need large flight muscles? If they exist in a world with magic, then a reasonable explanation is that they evolved a natural flight magic that offsets their mass and lets their wings be much weaker than otherwise required. Same explanation for the firebreathing - human mages study to recreate the same fire spells that dragons know genetically. $\endgroup$ Apr 7, 2019 at 12:12
  • $\begingroup$ The 2-legged/2-winged 4-limbed dragons (a.k.a. "wyverns") have been recorded since 752 AD, when mentioned as being on the banners of the Kingdom of Wessex (which itself existed from 519 AD until the 10th century - how early it was using a wyvern is unknown), and the Oriental-style 4-legged/0-wings 4-limbed Dragons are recorded from 323 BC in Mesopotamian art. So, I would argue that "most dragons before Skyrim and GoT" did not have 6 limbs, just the ones that you personally were most familiar with. Given how large China and India are, the wingless versions are probably the most prevalent $\endgroup$ Apr 23, 2019 at 10:37

4 Answers 4


Im not an avian biologists so i do not know if this configuration would be biomechanically sound, however it may prove to be some help to you.

A break down of the skin, muscle and bone layers of a dragon Credit to Christopher Stoll

Shown here is a break down of the skin, muscle and bone layers of Toothless from the How to Train Your Dragon series. This image clearly depicts how and where the bones connect and where the muscle attatches to the skeleton. As i mentioned, i am not an avian biologist so i do not know if this setup would actually be functional in the real world, you may very well be able to draw inspiration from this though.

Edit: As you specified for an image containing the pectoralis, i found another one that depicts it.

enter image description here http://mythicalanimalscience.blogspot.com/2015/04/dragons.html

  • 2
    $\begingroup$ @Mephistopheles Due to how it's positioned you cannot see it on the leg portion. It is hidden beneath what have to be the triceps as the attachment of the pectoralis is too high up. The weird "necktie" muscle is supposed to be the pectoralis major of the wings, but it's very small andhow it's positioned means it would move the front paws backwards and forwards with each wingbeat while losing some power. The second picture has the wingmuscles behind the front paws, which means the wings are farther behind and would have the dragon fly face downwards but it might work. $\endgroup$
    – Demigan
    Apr 6, 2019 at 19:59
  • 1
    $\begingroup$ Regarding picture #2 Thank's you just solved (almost) all of my problems! $\endgroup$ Apr 6, 2019 at 20:00
  • 4
    $\begingroup$ @LiamMorris not an unsurmountable problem, but the artist of the drawing was so focused on the wings that he forgot to give the front paws proper joints. It can only move the paws straight forwards, and is very limited in it's motion. Again this isn't insurmountable. The artist also forgot to make the lower ribs stronger, as right now the sternum would be pulled into the abdomen and wreak havoc with the organs there with each wingbeat. an elongated ribcage would make the dragon walk very stiffly but allow it to fly without commiting seppuku. $\endgroup$
    – Demigan
    Apr 6, 2019 at 20:13
  • 1
    $\begingroup$ Let us continue this discussion in chat. $\endgroup$
    – Demigan
    Apr 6, 2019 at 20:46
  • 1
    $\begingroup$ considering the first image basically has no flight muscles, no it cannot be sound. $\endgroup$
    – John
    Apr 6, 2019 at 21:34

Split the forelimb.

Making a 6 limbed creature from a tetrapod body plan is tricky. WB stack is littered with efforts on this front. Here is a new idea.

First, comparative skeletal anatomy.

comparative forelimb skeletal anatomy


Look at the whale. Now imagine splitting the distal forelimb into two limbs: one with the radius as core and the other with the ulna. The radius and associated digits (5 at baseline but you can add more digits; polydactyly is fine) becomes the robust wing, availing itself of the scapula and other support structures.

The ulna also has digits and is much less robust. It would not be a stocky limb to match the hindlimb but something more like a Tyrannosaurus. These small forelimbs would touch the ground and allow ambulation but for fast motion they are not the equal to the hind limbs.

Having powerful back legs, powerful wings and spindly forelimbs means these dragons would not look like Toothless or Smaug.

  1. Running at speed would be bipedal. The dragon would rear up and run like a bird. Wings might be used while running for propulsion or to jump. Some people think this use is how birds evolved wings in the first place.

  2. The small front limbs would lend themselves to more delicate manipulation. Dragons could sit up and have a smoke, or play cards.

  3. Front limbs might move during flight because they retain a connection to the humerus. Probably it would be some sort of rhythmic movement mirroring the wings.

  4. In general I think dragons are depicted as too robust. I picture this dragon as along the lines of a crane.

  • $\begingroup$ What if I made the bone larger, but hollow, and used my graphene-magic on it? $\endgroup$ Apr 6, 2019 at 19:45
  • $\begingroup$ All the bones should be hollow to reduce weight, bird style. Hollow long bones should not be too troublesome - mammals have hollow bones too, but use the space for bone marrow. Your dragon can make blood in its spleen and have air inside its hollow bones. $\endgroup$
    – Willk
    Apr 6, 2019 at 20:08
  • $\begingroup$ Your forelimb description made me think, Wasn't there a question awhile back asking how a dragon would go about knitting/sewing? Ha, there was! $\endgroup$ Apr 6, 2019 at 23:27
  • 1
    $\begingroup$ One problem with this proposal is the lack of necessary musculature. In terrestrial vertebrates, both bones are operated by the same set of muscles. In your dragon, you'd have to add new muscles, as well as rework the elbow entirely because you want the radius and ulna to face opposite each other rather than parallel. I do like the idea though. $\endgroup$ Apr 7, 2019 at 12:08
  • 1
    $\begingroup$ @Willk nice idea but you don't have to assume that the ulna isn't robust enough; sizing anything up or down is pretty easy for evolution, it's merely a question of how long the growth phase is allowed. Just look at the images you provided yourself for examples :-) $\endgroup$
    – toolforger
    Apr 7, 2019 at 16:39

The problem with wings is the necessary amount of power to lift the creature.

This is a bird skeleton:

This is a bird skeleton

Note the size of the sternum (the bone that all your ribs connect to). It's massive in comparison to ours and stands far out of the chest while a human sternum is basically flat in comparison. This sternum is what your pectoralis major muscles are attached to (the chest muscles).

The sheer size is because the muscles attached to it, shown here:

bird muscles

These absolutely humongous muscles are required to lift the weight of the entire creature up in the air, and the larger the bird the more % of the bird must be pectoralis major muscles just to keep it in the air due to the square cube law, but you can ignore this for the sake of cool. Despite this suspension of belief on the part of the square cube law you still want a bodyplan that can handle 4 paws and 2 wings simultaneously.

Note how these skeletons lack a large scapula. In the first picture you can see it mentioned but it's tiny and largely immobilized because everything is focused on that up/down movement of the arms, not about reaching forwards, backwards, upwards or downwards.

Having to have muscles attached to the chest for these extra appandages would diminish the amount of muscles for your wings. You could slightly circumvent this through kinematic chains.

An example of a kinematic chain is your quadriceps of your leg, or any other muscle that moves over more than 1 joint in the body. If you have muscles pull on one end of the bone they are attached to, you can use that to pull on the quadriceps, the quadriceps us that to pull on the muscle one joint removed, allowing you to transfer the muscle power from one bodypart to another. You attach a portion of the pectoralis major muscles to the dragon's leg, and have muscles above it attached to the wing (likely an adapted version of the triceps, romboideus, one part of the deloideus and trapezius). When flying the leg will be pulled down in the same motion as the wings and the muscles above that will simultaneously pull on the wings, allowing you to transfer the muscle power that went into the leg into the wing as well. This isn't wildly efficient and would likely make the leg flop up and down with the wingbeats, but it's an option. Ofcourse if you time it right and pull just as hard on the leg as on the wings, the leg would effectively keep still during flight. Although you'd be better off pulling it in as it would stretch the muscles above it and with that pull the wings.

  • $\begingroup$ Although note in birds that swelling is the muscles to flap and the muscles to lift the limbs, birds layer thos muscles on top of each other. $\endgroup$
    – John
    Apr 6, 2019 at 21:30

Any depiction by Larry Elmore should suffice. More like angles; their forelegs are actually arms, and the wings protrude from their back, beginning just below the shoulder blades.

enter image description here
Dragonlance's blue dragon, Skie, as pictured by Larry Elmore in the original cover art for, Dragons of Winter Night.

  • $\begingroup$ It's a nice picture, and typical Fantasy depiction... but from an anatomy point of view it is plainly insufficient. Wings typically require a massive muscle to beat them (it has to lift the whole body, after all), and it's very unclear in your picture where said muscle would be. Actually, it looks like there's no such muscle and the wings are just aesthetic but non-functional. $\endgroup$ Apr 7, 2019 at 9:24

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .