Saturday, 5 May 2012

A century of thoats

As post titles go, the one above may be a bad choice: it will probably only catch the attention of those who already know what a thoat is. To attract more readers I should perhaps alter the title; the second-highest rating post in this blog, as far as numbers of viewers is concerned, had 'Avatar' in the title, so I should probably learn from that. So here is an alternate title: 'Why do animals such as the thoat in "John Carter of Mars" and the thanator in "Avatar" walk in such an illogical manner?

A thoat is a large eight-legged Barsoomian animal, 'Barsoom' being the native name for the planet Mars in Edgar Rice Burrough's works. Burroughs starting writing Barsoom novels around 1912. I never read any, for the simple reason that there were no Barsoom books in translation around when I was at the right age to enjoy them. I did read his Tarzan books, though, and suppose that I would have read Barsoom novels with equal appetite. The reason Barsoom caught my attention after having ignored it so far, lies in the recent movie 'John Carter of Mars' (JCoM).

Click to enlarge; copyright Michael Kutsche

The wonderful image above is from the DeviantArt account of Michael Kutsche, who designed the thoat for JCoM; the thoat is the eight-legged beast, not the individual sitting on top. Have a look at how Kutsche approached the 'leg problem': where do you put eight legs without them looking odd, if not altogether ridiculous? Kutsche solved the problem by dividing the eight legs in front and hind groups. In build, the legs look a lot like those of a large mammal such as a rhinoceros or an elephant. In fact, the original hind and front legs were simply copied and pasted as close as possible to the original. Where there was one leg there are now two, moving in unison. Being so close together, that is all they can do, as otherwise they would knock into one another. This solution is very reminiscent of the large animals in Avatar, except for the fact that these had six legs, with just the front legs doubled. I criticised that arrangement in a previous post, thinking that it did not make much sense from a biological point of view. It felt it would be easy to learn how six-legged animals could or should move: after all, Earth is literally crawling with insects. For eight-legged creatures you can look at spiders or crabs.

The video fragments above shows a few seconds of thoats walking in JCoM (sorry about the quality), illustrating the 'doubled leg design'. Bt the way, I wonder why the legs are so immensely thick. Their thickness looks about right for an athletic elephant-sized mammal on Earth with just four legs. But with eight legs, each leg can be more slender than if there are four (read here for posts on leg design, gravity and the number of legs). As it is, the animal looks as if it was designed for a planet with a very high gravity, not for Mars, where gravity is just 38% of that of Earth.

Click to enlarge

The Barsoom fictional universe has been around for a century, so many illustrators over the years must have faced the problem how to design a plausible as well as dramatic eight-legged big animal? The drawing above, by John Allen St John, must be one of the first depictions of the Barsoom universe (I found it here). The style of the drawing fits the early twentieth century. This early thoat does not show doubling of the front and hind pairs of feet, but shows tripling of the front feet, keeping one pair of hind legs. There seems to be a generic 'copy and paste legs' solution operating here.

In the late thirties there was an attempt to produce an animation of JCoM. Only a bit can be found now, coming from ERBzine, a very large website on Edgar Rice Burroughs. I woulld like to include it here but am having upload problems; I may rectify that ater, but for now, here is the YouTube version. Look at how the thoat lands on its feet: there is a definite phase offset between the successive pairs of feet, resulting in a jumping gallop. This is one of the very few thoat designs with phase differences between successive pairs of legs.

Click to enlarge

Edgar Rice Burroughs son, John Coleman Burroughs, worked on his father's creations. He produced a wooden thoat model, shown above. I found these images on the Erbzine site. It is a pity that the model cannot be seen better. The legs are so close together on the body, that they must all move in unison on one side of the body, or else they will knock into one another.

Click to enlarge; copyright Frank Frazetta (I presume)

The late Frank Frazetta, famous for his equally but not similarly well-developed heroes and heroines, produced quite a few Barsoom paintings. I found no thoat among them, but there are two banths, Barsoomian lions with ten legs. The ink drawing above shows a banth from the front, a nice trick to avoid looking too closely at its shape. The legs on one side seem to be moving in unison. For the oil painting, Frazetta chose to hide most of the banth's legs from sight. For a painter of Frazetta's skill this is probably no coincidence; did he feel that showing all legs would not work?

Click to enlarge; copyright William Stout

The image above is by William Stout, a well-known painter of palaeontological scenes. His thoat is livelier than most, as can be expected from a Stout design. And the legs? Well, a doubling design again.


So, what can be concluded from studying a century of thoat design? Some of the best fantasy illustrators worked on the problem, and many gravitated towards the same solution: they doubled or tripled pairs of legs, and as such did not depart too much from the familiar mammal pattern. To me the 'doubling solution' does not seem like a good idea, but if all these wonderful illustrators chose it, I may have been too harsh in my judgement. A good artist or illustrator will have learned human and some mammal anatomy, but that education is not likely to include hexapod and octapod locomotion. In fact, the necessary knowledge is not that easy to find, and has not been adapted for non-specialist use.

Copyright Gert van Dijk

What would you get if you do take such knowledge into account? Well, probably something as in the animation above (adapted from a model designed to study Furahan rusp locomotion). I made no effort to define the body and the legs all have the same shape. The main point was to show an eight-legged gait that can work well. The gait is based on one described in a scientific paper on spidr gaits, and concerns a slow walk. I applied that gait to a model in which the legs are held vertically rather then horizontally, making the design better suited for large animals. Personally, I think it makes more sense than the doubling design.

You could argue that using such a true octapod gait or the simpler 'doubled leg gait' does not matter, as the audience will probably accept doubled legs as easily as a more sensible gait. Maybe; the audience may also feel that the true octapod gait has a more alien feel to it, and that should please the designers. All in all, I probably overestimated how accessible the required biological knowledge is. Well, that is a good reason to continue this blog. What else could be done with it? Phone Hollywood, perhaps?...


trex841 said...

you say octopod gait, and i see the megasquid. pretending for a moment the creature could actually hold it's self up, they said that it had a walking pattern for 8 legs. would that work?

Sigmund Nastrazzurro said...

Trex841: If its leg would keep it upright then you would have an octapod. I do not remember its gait and cannot check the video now. I seem to remember a slow movement with only one or two legs off the ground at any time; this might be to prevent loading the legs too much, or just an adaptation for enormous size.
Given suitable legs I see no biomechanical reason why it would not work (note I am not talking about other arguments against terrestrial cephalopods).

mithril said...

the "doubled quadruped" approach is older than dirt. for example, the eightlegged horse Sleipnir, son of Loki and the horse ridden by Odin (a complicated, and a bit wierd, story; )is one of the few creatures we have specific art for from the medieval age. the Norse, and the cultures predating them, depicted Sliepnir as basically being the same as a regular hourse, but with twice as many legs for and aft. their art usually depicted, (in it's very simplisitc form) that it used the gait of a regular horse, the extra legs moving in unison.

this depiction certainly was a case of trying to use a familiar object to describe a fantastical one. i wonder how much of the tendancy to do similar with more recent fictional creatures comes from "doing the reasearch", finding sliepnir, and deciding if it worked for the norse it works today?

Spugpow said...

The sabbatical seems to have done its job: this post is fascinating!

I'm also a little mystified by designers' inability to come up with plausible gaits for their multi-legged creatures. I remember figuring out the gaits of insects and spiders when I was little by watching them from below through the bottoms of glass jars, so presumably creature designers could do the same.

Then again, challenge of animating an eight-legged animal with a realistic gait must be considerable: not only will it walk and run, but also change direction, shift from foot to foot, and make false starts. This might explain the preference for paired legs.

The Megasquid walks by moving the two outer legs on one side in unison with the two inner legs on the other. It's still "twinning" its legs, but this time in imitation of an insect's tripod gait.

Spugpow said...

This lends some credence to your idea, Mithril:

Sigmund Nastrazzurro said...

Mithril: I had limited my research to the direct ancestors of JCoM, and you are right to point out that there quite a few odd mythological creature with unusual numbers of limbs.
But I wonder whether a modern artist faced with the problem of depicting eight legs realistically would find the Scandinavian depiction of Sleipnir a good example. Mind you, if I had to paint Sleipnir, I too would definitely use the doubling approach. Why? Because Sleipnir is a horse with eight legs! It should look like a horse, even if the design makes little biological sense.
But if you wish to design an alien animal, it should not look like an eight-legged horse. Apparently that is more difficult than it looks.

Spugpow: thank you. A suitable gait for octapod animals can certainly involve moving legs in unison, as you must have found out looking at insect gaits as a boy. The thing is that there are reasons why specific legs move together.

Petr said...

So far it seems creature creators think that leg grouping is the only possibility:

In the creature below, the "lobe-fins" on the back are also limbs, so it has the same bauplan as the thoat or sleipnir:

And here are grouped hindlegs for a change:

Evan Black said...

An important thing to keep in mind when considering the plausibility of a creature is the motivations of the artist/speculator. In some instances, a creature's design is based more on aesthetic effect than on evolutionary plausibility or biomechanical sense, and there are several factors that influence such a decision.

Firstly, artists are typically more interested in evoking a visceral, instinctive association in their audiences than investing research in biomechanics. They want creatures like thoats or thanators to be instantly recognized as bovine/equine/pachyderm riding beasts and giant panther-like predators, respectively. They’re using metaphorical imagery as creatively as writers once did by referring to a lighter-than-air vehicle as an “airship,” or a spacegoing vessel as a “spaceship.” (That reference is so widespread that the word “ship” has long since been expanded to include vehicles traveling in media other than water, and the term has lost much of its metaphorical reference.) So, to respond to creations that are based on this criterion is like acting shocked and disgusted when hearing that someone has “cast their eyes heavenward.” It’s simply not appropriate to assume that a human being can literally do so without removing his or her eyes and tossing them into the air, and it’s important when looking at speculative or xenobiological artwork, even that which claims to be evolutionarily plausible, to separate out the biology from the artistry.

Secondly, artists are usually aiming for internal consistency rather than adherence to biophysical or evolutionary likelihoods; this internal consistency often depends on the genre of the story as well. In a pulp story—let’s use the Barsoom series as an example—where an earthman is transported to a world where the only adjustment he has to make in order to survive on Mars is to learn how to walk in lighter gravity without tripping, where Lowellian canals crisscross the planet, and where starships are able to negate gravity by riding on light waves, it seems inappropriate to expect a biology that doesn’t delve into the fantastic. We might as well consider the evolutionary plausibility of the balrog.

I suspect that a similar process took place in the conceptualization of Avatar’s creatures. After receiving design details and instructions from James Cameron, concept artists create imagery to bring that imagined world into light. It’s apparent that designs for final production were chosen more based on the rule of cool than on plausibility, but I think it’s commendable that the designers gave attention to making an evolutionary connection between hexapod flyers and tetrapod Na’vi. Plausible or not, I think that the “twinning” of the foremost limbs seen in so many creatures is a reflection of this effort to create at the very least an internal logic. I’ve seen the same attempts to force disparate concepts into evolutionary cohesion in countless SpecEvo projects (and could potentially be accused of doing so myself).

Finally, artists often deal with time as a factor. I’ve been working on my xenobiology project for three years now, and while I’ve produced a great deal of material, it’s largely been designing for its own sake, and not intended to fit with any particular story (yet). But that’s three years of learning and research and experience that, frankly, concept artists typically don’t have available to spend on a single commission, so they fall back on what they’ve studies for perhaps decades: anatomical studies of humans and animals. Given a time crunch, you draw what you know.

So yes, we should point out the implausibilities we see in Hollywood designs, and discuss what would improve the biomechanics of given creatures, but we shouldn’t be surprised when biological plausibility fall short on the silver screen. To expect otherwise is to look for an orange flavored apple.

Evan Black said...

Man, as I read over that last post of mine it sounds really angry. I apologize if anyone took it to be harsh, which was not my intention.

Sigmund Nastrazzurro said...


I'll say I was a bit surprised about your rant, but I do not think anyone among the readers would have taken it as a personal rebuke, and do not think anyone will have been offended.

The points you raised have come up in this blog more than once. On the one hand, we, i.e.. the writers and amateurs of speculative biology, know that the creatures in novels and particularly films could be made more authentic, more biologically plausible, with rather limited investments. On the other hand, we see from their products that film makers do not care one hoot about plausibility at all.
Mind you, for them a 'lack of plausibility' may only mean a problems in the scenario that causes lower sales. Do not get me started on historic accuracy, something much easier to check than biological accuracy...

In my Avatar post I supposed that the film makers may have preferred the dumb solution (doubling of limbs) over a smarter solution (a proper hexapod gait) because they felt that the dumb solution resulted in a bigger resemblance with Earth animals that would help the audience relate to the animals more easily. In my last post, I wondered whether the advisors may simply have failed to achieve the biologically more sound solution, and hence failed to make the most of an opportunity. I supposed that a true alien gait would have been appreciated as such by the audience, something that would translate into more tickets.

So, what is going on? A calculated dismissal of knowledge, implying contempt of the audience, or are we dealing with simple ignorance? I fear the former is often true, but hope for the latter.

Evan Black said...

I'm glad to see that my points managed to find their way through my surprisingly ranty post. I still don't know why it came across so intense. Perhaps it was something I had eaten. :) I've held off on posting this comment until now so that I could make sure it didn't have as much intense language in it to clog up what I'm actually trying to say.

I think an argument can be made for simple ignorance on the part of designers. After all, how many film makers and conceptual artists can boast a Ph.D in biophysics, astrobiology, or any other relevant sciences? So much scientific understanding in the general populace is, unfortunately, based on "Hollywood science." Even within the small niche of enthusiasts for xenobiology and speculative evolution, wrong science pervades. I don't know how many interesting ideas I've explored for aliens that I later had to throw away because things just don't work that way.

A calculated dismissal of knowledge could be at work here as well, but I don't think that it's necessarily connected with a contempt for the audience. Back in February we were able to recognize the implausibilities of the sandworms of Dune without labeling Herbert as some kind of shyster, but rather that his focus was on something other than the biology of a creature that, despite its role in the plot, is really only present as a symbol and to evoke the sense of wonder that is a staple of Science Fiction. The same could be said for Barsoom, which is based on a book that could never claim to be 'hard' Science Fiction.

Yet you mention that the efforts of almost a century of artists prompts you to consider your dismissal of the 'doubling solution' as "harsh." Why question your own findings in the face of evidence created by artists? It makes me wonder why this has been the instinctively preferred method of illustration for so many decades; is it some bias for the symbolic power of familiar images, like how the quintessential design of the dragon has been so popularized in world culture that it's very hard to deviate from the two major types (Asian and European)?

It also makes me wonder about the claims made about Avatar's plausibility. Didn't Cameron say he had several scientists weigh in on the many aspects of the the movie,including biology? I wonder if they have any evidence to support the plausibility of the leg doubling, or if this was something that simply slipped through their scrutiny and allowed to follow the cultural preference discussed in this blog post.

Evan Black said...

Too harsh again? :(

Sigmund Nastrazzurro said...

Not at all! I was just busy; with repainting the seasoar, among other things. I will answer later (and will also send you the repainted seasoar in private to hear your opinion).

Christian C. said...

@ Sigmund

If found this article very interesting, and I think the reason the thoats had their legs doubled in movement is because they need to run fast and not just walk. By the way, did you post any other articles about large animals having multiple legs, because I'm looking for facts to incorporate in a sci-fi novel that I'm currently brain storming. I'm thinking about 10-12 legged organisms for inhabiting the planet GJ 667Cc.

@ Evan Black

You mentioned that you didn't get how the Na'vi from Avatar had four limbs when the other life forms had six. I believe they tried to explain that with another organism that was featured in the movie, the prolemuris. They are to the Na'vi what chimps are to us. If you have a good look at them you will notice several things. Their arms are fused up to the elbow to allow faster movement in the trees and each of their forelimbs have two fingers, indicated that with further evolution the arms would fuse all together and result in two arms with four fingers on each hand. Also, their heads have some of those head nerve extensions similar to the Na'vi's "hair" and a single large nervous extension on the backs of their heads. Hope this helps.

Evan Black said...

Actually, I pointed to that very relationship with Prolemuris as a commendable effort on the part of the designers. While I question the evolutionary need for the merging of four limbs into two in the first place (for example, how does a bifurcated limb allow for faster movement in the trees?), I do recognize that the discrepancy between the four-limbed and two-limbed clades was a decision based on the vagaries of filmmaking and not on xenobiology and applaud the attempt to provide some kind of explanation.

Anonymous said...

>I think the reason the thoats had their legs doubled in movement is because they need to run fast and not just walk.

But then they need to avoid tripping over their own feet, or stepping on their own heels. much harder to do at speeds of "jog" or above.

Petr said...

@Rodlox - spiders and crabs can be fast despite having no gap between the second and third pair of legs.
Thoats are just an example "how to make it familiar enough but still different" logic, which is why the heads of the animals look very earthly. They just made it so even the movements can be familiar to the audience.
Also, given that the thoats have eight legs on Mars and the legs are still as thick as of the four-legged terrestrial rhinoceros, I doubt a lot of biology went into it. If you have a multiple-legged animal in weak gravity, the legs can be slender and still carry the animal with ease, right? =D

Sigmund Nastrazzurro said...

Evan: I do not think we disagree, but I may have not stated my meaning very clearly. You wrote that an argument can be made for ignorance on the part of the designers, and I made a point that perhaps it is a bit much to expect an artist to know about six- or eight-legged gaits. While irritated at first, now I am inclined to forgive people for lacking such very specialised knowledge.
What I wonder about now is how directors etc. might respond when faced with biologically better solution. For me, that also means more alien, but would they agree that more alien is better, or would they prefer to stick to Earth-with-a-twist?

Christian, you wrote the following: "thoats had their legs doubled in movement is because they need to run fast and not just walk". There is an assumption here that having more legs allows an animal to move faster. That is not as straightforward as that. In fact, there are clues that the opposite may be true. Crabs and insects may in fact use fewer legs when running than when walking slowly. Off the top of my head, cockroaches may even use only two legs when running, as do some crabs!
I only wrote one post dealing with the number of legs: . Please read the responses as well. By the way, I have been working on my 'rusp' concept and may write a post about them. I see no problem with many-legged large animals, but I would refrain from them being fast; no running, and probably no jumping at all.

As Rodlox and Petr pointed out, if you design an animal with multiple legs, make certain that their thickness is in line with their number, the size of the animal and local gravity, and that their gait makes them useful without making them stumble.

As for limbs fusing, I find that an extremely unlikely concept. I can envisage limbs shrinking and disappearing, but not fusing. Wouldn't the result (such as the prolemurs arms') be very impractical? (I wrote a bit more on that in my Avatar post: )

Evan Black said...

Whether or not directors would prefer a more accurate but alien gait over "Earth with a twist" is a matter of taste, I think. And I think that's what I was trying to get at (and failed) in my other posts. For those of us who revel in observing aliens in motion, the choice is obvious, but for some filmmakers the emphasis is on whether or not the audience will identify with the creature (hence the humanesque Na'vi). So convincing a filmmaker to go for the more accurate but more alien design might be an uphill battle in some cases, and may require mockups or animations that can show just how majestic, intoxicating, and dynamic a more accurate gait can be. Things like your "Walking With Hexapods" page does a good job of communicating the wonder of gaits beyond four legs.

I also agree that limb atrophy is more likely than limb fusing. Fused limbs happen, but I can't think of a scenario where a) such a fusion would prove to be an evolutionary advantage, or b) where a series of transitions between two separated limbs to two unified, functional limbs could provide a series of advantages. I mean, I can see how the physiology would select for a less awkward and impractical set up than the bifurcation we see in Prolemuris, but what mechanism is prompting this fusion in the first place that can't be better solved with atrophy? I can't think of any. Unless some compelling evidence presents itself, I only applaud Avatar's designers for their attempts at a sound explanation of the transition leading to the Na'vi.

Petr said...

What I think is interesting, that fused digits happen, in contrast to fused limbs.
here is a mule-foot pig:
Here is an obvious choice, the chameleon:
And here is a marsupial foot, which was a rather surprising find for me:,r:0,s:0,i:71
Can you see the suspicious "digit" with two claws on the tip?

what is interesting, though, is that both, limbs and digits show webbing in many cases, but only digits can fuse.

It might be that the bones within one limb are closer together and therefore more likely to fuse than bones of two different limbs, so if you "need" to go from a higher number of limbs to a lower one, reduction is the only way how such thing can happen.

back to the "leg grouping" problem. I am surprised no one pinted up certain mites and praying mantises. Some mites clearly show "two groups of four" arrangement, and mantis freed the forelegs from the other legs, so there actually IS an example of "legg grouping" in the natural world.
I came to a conclusion that with a crawling animal, the basal joints of different pairs of legs can be placed close to each other, because the feet are far apart and therefore the legs can still move - the longer the legs, the room for movement - so I think that "leg grouping" is forgivable in a crawling animal. For an animal with upright legs, however, basal joints close to each other mean that the feet are close to each other too, so the limbs have to move in synchrony - Here I can lively imagine how the "rear foreleg" bruises the "fron foreleg" very easily. I think that even if the thoats came from a crawling animal with grouped legs, the legs would most likely redistribute evenly along the body to grant the legs the free space they need.
Also, I don't think "limb grouping" is a big problem for an aquatic animal, some fish have their pectoral and ventral fins very close to each other, and it doesn't seem to cause them any difficulty. ;)

Anonymous said...

Hmm, but all the multi-legged creatures on Earth are arthopod invertebrates. I think leg-doubling makes sense if trying to go for a 'mammalian' anatomy.

Also, front legs are connected to the ribcage and hind legs to the pelvis. If a creature were to have eight legs, would two pairs be attached to the ribcage and two to the pelvis? I think even spacing would put the middle legs at the abdominal flank, an impossible arrangement due to lack of belly bones to support the limb girdles.

taotie said...

Looked for the discussion of bifurcated limbs on this blog because of the Gibbets from the Serina project, and for some reason the Gibbets seem to be the complete inverse of the Prolemuris?

The Prolemuris, as mentioned, has a single upper arm with fused humerus pairs, but splits ath the elbow. This was heavily criticised since it seemed unlikely that a partly-fused limb offers any advantage.

But then we have the Gibbet:

As seen in the image, the Gibbet is one of several species of arboreal fish, with fins modified into long-fingered hands for climbing. In the Gibbet, however, the arm and forearm became shortened, while two fingers on each hand became hypertrophied and eventually developed little fingers of their own: again, two pair of partly fused arms like the Prolemuris, but from a single pair split into two.

So in redesigning the Prolemuris, perhaps the common ancestor of Na'vi and Prolemuris lost one pair of limbs via atrophy, and the Prolemuris is actually evolving back four arms, by splitting its two front limbs with shorter arms and longer "arm-fingers"...