How many legs should animals in speculative biology projects have?

By Sigmund Nastrarruzzo and Biblaridion

We recently published the first of two posts in response to comments on Biblaridion’s YouTube review of Sigmund/Gert’s book ‘Wildlife on the planet Furaha’. The number of legs of Furahan creatures evoked many comments, in particular Furahan scalates. Scalates are animals with some similarities to Earth vertebrates, such as a comparable size range and bilateral symmetry. But the six legs of Scalates urged some people to argue that larger animals should always have at most four legs. The previous post dealt with limits posed by a Bauplan: the genetic building plan that defines an animal’s main anatomical features. In this post, we start with discussing Bauplan effects some more and then try to work out what effects the number of legs may have on animals. 

Bauplan limitations 

The odd thing about a Bauplan is that its contents are largely locked and cannot be modified through evolution, and yet that same Bauplan can only have come into being because its contents were at one point highly malleable! If the number of legs is locked in a Bauplan, that number stays fixed. The four limbs of Earth vertebrates are locked, so vertebrates will not have offspring in which the basic pattern jumps to two, six, or eight legs. There are three major lessons here. 

1. The fact that ‘four legs’ is part of the vertebrate Bauplan does not mean that this number works significantly better or worse than another number; all we can conclude from the long history of vertebrates is that four legs work well enough. 
2. The number of legs does not have to be locked in a Bauplan in each and every animal clade! The number is in fact variable in Earth millipedes, and speculative biology creators can use that fact to decide whether the number of legs in their creations is fixed or malleable (it is variable in Furahan rusps). 
3. The basic instruction ‘make four limbs’ still allows anatomical modification of those four limbs. Evolution can find other purposes than walking for front legs, a principle baptised ‘centaurism’ in this blog; examples are the front legs of mantises, wings in birds, the hands of primates and theropods, and mouth parts in just about any arthropod. The result is that the animal gains new functionality at the cost of a pair of walking legs. Such centauric evolutionary changes probably starts with animals using their front legs both for walking and for a new purpose, and over time the new function takes over completely, so the walking role is lost. Some commenters on Biblaridion’s video felt that many-legged animals would lose legs until they arrived at four legs, without a new function. But in an evolutionary process each step has to bring an advantage. To make their scenario work, animals would have to start using only four of their six legs even though all six were fully functional; we see no clear advantage in that, and evolution is not driven by future goals, but by advantages here and now. 

Building costs 

If a leg is only a weight-bearing cylinder, then each leg can be more slender the more legs there are. But what is the total mass of many slender legs, compared to that of a few stout ones? The calculations on that question (here and here) showed less total weight with fewer legs, with the least mass for just one leg! But you only have to think about how fatiguing it is to hop on one leg to realise that building costs cannot be the only consideration. In evolution, many factors usually play a role, and the best solution then is a compromise between conflicting demands. For example, factors such as stability and surviving harm may outweigh saving weight. 

Surviving harm

If a one-legged or two-legged animal breaks a leg, it will be unable to eat or drink, and will quickly become someone’s dinner. Four-legged animals are probably also doomed (dogs and cats can adapt in amazing ways, probably because humans help them through the critical phase). A six-legged animal can probably limp away after the loss of one leg, and the consequences of injury become less severe with a larger number of legs. Do millipedes even walk slower after losing a leg? In short, multiple legs allow better chances of surviving injury than four or fewer legs. 

Speed 

Animals that walk fast use four principles to speed up. The first three are simple: steps become longer, step frequency increases, and the fraction of time a leg is in the air increases, while time on the ground decreases. The fourth factor is gait, meaning phase differences between legs. During a slow walk, the body has to be supported at all times, which is best done by having all legs move at different times, so no two legs move in synchrony. For slow walks, this works best with four or more legs. However, fast running works best if the gait contains jumps, meaning periods in which all legs are off the ground at the same time. To achieve that, leg movements have to be synchronised to a degree, which can -in principle- be done regardless of the number of legs. For instance, take a Furahan rusp with 24 legs. When such animals run, all right-sided legs move in unison, and so do the left-sided legs, but exactly 50% out of phase. 

Bipedal running cockroach; click to enlarge. From Full and Tu 1991

 

However, that is not how many-legged animals run on Earth. Cockroaches, crabs and lizards may use just two legs when running fast. Why do they not use all their legs and run like rusps? Well, rusps have their feet close to the midline, which makes it easier to support the body with legs on one side only; cockroaches, crabs and lizards have splayed legs, which makes one-side support difficult. Another part of the answer appears to be that the preferred legs for bipedal running in cockroaches and lizards are longer than other legs, so the other legs are less useful. Note that, even though crabs may need only two legs for maximum speed, this hasn’t resulted in crabs with only two legs! The other legs are useful at other speeds. For instance, crabs need good stability when walking slowly, helped by having more legs, and their life may depend on their ability to cling to a rock using all legs in heavy surf. 

For large animals, it is less likely that one pair of legs provides enough power for maximum speed, so they need more legs to achieve that. Perhaps another factor weighs in too, such as the animal not achieving stability with only two legs while running. 

Stability and size 

We will use ‘stability’ essentially as the likelihood of not falling, while standing or walking. Animal size is extremely important here, because the effects of gravity are relatively much more important for large than for small animals. In contrast, sideways forces such as wind and water currents are more important for small animals.  

Gravity always presses large animals firmly to the ground, so their feet do not have to cling to the ground or grasp it. Stability primarily means keeping the centre of gravity over the area defined by where the feet touch the ground (a support diagram), and secondarily being able to keep the body orientation the same. 

Small animals, say insect size, do need to cling to or grasp whichever surface they reach, regardless of whether that is vertical (as in the illustrations below), horizontal or even upside-down. For them, stability is keeping the body position and orientation stable regardless of where the footholds are. 

What happens to stability and position for large and small animals when the number or legs increases?

Click to enlarge; copyright Gert van Dijk

• One leg allows a large animal to stand and to hop, with gravity ensuring it always returns to the ground. One-legged standing and hopping amounts to a continuous balancing act, which in turn needs very sophisticated nervous and muscular systems. Animals that are just learning to walk on dry land will not yet have evolved such sophisticated systems, so it is doubtful that one-legged animals could even begin to evolve the skill of walking. For a one-legged small animal, hanging from its one leg, life is simple: if it relinquishes its grip, it falls away from its support surface; not a good idea. 

Click to enlarge; copyright Gert van Dijk

• Two legs means fewer directions to fall in, which helps. Birds, other dinosaurs and people all walk well with two legs. It is not easy though: faster walking requires dynamic stability, meaning the centre of gravity needs to kept over a continuously changing support diagram. To do that, you will again need exquisite control, so we advise bipedal animals that aim to become terrestrial walkers to not bother. Very small two-legged animals could theoretically hang from one leg while moving the other, but it would be very difficult to prevent the body from just hanging down from one leg: not a good idea. 

 

Click to enlarge; copyright Gert van Dijk

• Three legs, for larger animals, allow a neat support triangle; that is all you need to stand still, and you do not need complex control for that either. But walking with three legs will require excellent control. Very small animals can hang from two legs, but it will still be difficult to control body position. 

 

Click to enlarge; copyright Gert van Dijk

• Four legs, in larger animals, do not make standing easier than with three legs, except when there are strong sideways forces such as water currents or strong winds. If there are, it helps to have a leg sticking out in the direction of that force. Walking with four legs offers a fundamental advantage over walking with three legs: you can lift one leg and still remain stably supported by the other three. This is still not exactly easy: tortoises, moving slowly, have stability problems, as they tend to fall in the direction of the lifted leg. It takes a specific order of moving the legs to prevent that, so here’s that neural control again. Small animals with four legs can maintain their body position using three legs, while one limb moves to a new foothold. 

 

Click to enlarge; copyright Gert van Dijk

Click to enlarge; copyright Gert van Dijk

• Five and more legs will not improve stability much, for large standing or walking animals. This probably holds for small animals too: with many legs, they can choose which ones grab a surface and which ones move to the next foothold, all while nicely keeping the body in position. However, many legs may help to grasp objects better: the eight legs of crabs may act as fingers, making a crab a sort of hand to grab rocks with; insects walking uphill may keep more feet on the ground than they do on a horizontal surface; the many legs of millipedes may help them bulldoze through dead leaves or soil. 

Stability becomes easier with more legs, at least up to a point; we guess that stability will not increase much beyond, say, six or eight legs. We suspect that ‘beginner walkers’, animals that are just learning the art of moving on dry land, will find the task easier if they have more legs, say six or more. If you are just learning to walk and have only four legs, don’t expect to be able to lift the body at all times at first! Start by crawling on your belly, and only try to lift the body when your control system can guarantee stability. However, if you start with six or more legs, you’ll learn to walk in no time! 

Planetary effects 

Many of these effects will depend on local gravity. On a planet with low gravity all legs can be spindly, and falling won’t hurt as easily. With a very strong gravity, legs have to be very strong and as many legs should be on the ground for as long as possible, which is easier the more legs there are. With a high gravity a lifestyle clinging to branches and twigs is asking for trouble, unless there are always lots of legs keeping a secure grip. 

Conclusion 

The number of legs plays a role in stability, building costs, surviving harm, locomotor efficacy and probably factors we didn’t consider. The number is restrained by evolutionary aspects and may be locked genetically. Whenever there are many factors determining an outcome in biology, it is usually impossible to identify one factor as the sole determinant of success or failure. We think that this holds for the number of legs too. The optimum number of legs is like most biological optima: a summation of many functions with lots of compromises. 

However, this can only be true if the number of legs can in fact vary freely, in which case evolution can modify the number as the outcome of such a summation. But if the number is locked, which seems more likely, then evolution has to work with what it was given. The number may then constrain evolution: for instance, an animal with just two legs would find it difficult to combine flying with walking. 

With all these factors weighing in, we do not think it likely that all large animals in the universe must have four legs (or six, for that matter). There are too many variables at play, so we expect the usual evolutionary mixture of Bauplan constraints and many factors weighing in. That seems to be how biology works, and for that reason speculative biology should work that way too. Are we certain of that? No, we aren't; it’s all speculation, remember? 

 

Short reading list

• Büschges, Ache. Motor control on the move: from insights in insects to general mechanisms. Physiol Rev 2025; 105: 975–1031
• Full RJ, Tu MS. Mechanics of a rapid running insect: two-, four-and six-legged locomotion. Journal of Experimental Biology. 1991; 156: 215-231
  
• R McNeill Alexander. Optima for animals, Revised edition. Princeton 1996 
• R McNeill Alexander. Principles of animal locomotion, Princeton, 2003
• Riiska CA, Harrison JS, Thompson RD, Nina JQ, Gallice GR, Rieser JM, Bhamla S. Katydids shift to higher-stability gaits when climbing inclined substrates. Integrative and Comparative Biology. 2025 Dec;65(6):1667-77. 
• Weihmann T. The smooth transition from many-legged to bipedal locomotion—Gradual leg force reduction and its impact on total ground reaction forces, body dynamics and gait transitions. Frontiers in Bioengineering and Biotechnology. 2022; 9: 769684  

What -on Earth- is this?

A week ago Jean-Sébastien Steyer, a French palaeontologiost and author of various books on speculative biology (here, here and here), alerted me to an Instagram video. The video showed a ghostly white arthropod against a black background, swimming vertically in a most peculiar way. I was immediately intrigued and at first wondered whether the video was AI trickery, but a commenter had already checked that and said it was real. I checked for myself (that's what scientists do) and found a key paper on Bathyopsurus nybelini, as well as a paper on another Bathyopsurus species.

 

Here is one of the videos you can find on the internet. Others can be found here. The one above was clear enough to show how the animal moved. Bathyopsurus nybelini is an isopod, the clade that also contains woodlice. The species is not very large, with body lengths of two measured specimens at 34 and 35 mm. 

The paper by Peoples et al., from 2024, makes it clear that the animal swims vertically or parallel to the sea floor, holding a bit of sargassum weed with its mouth parts. The swimming legs are at the rear end of the animal, which is at the top of the video. The animal is swimming backwards, but for a blind animal that may not matter much. The paper focused on the importance of bits of sargassum sinking and fuelling deep sea ecology. The authors had this to say about locomotion: Isopods swam paddle stroking above the seafloor with their two natatory legs (pereopods V VI) . Elsewhere they wrote: The large dual paddles support the isopod s characteristic swimming gait, which involves alternating power strokes of adjacent appendages. In addition to the broadening and separating of the two paddles on each appendage during the power stroke, setae provide additional surface area for propulsion power . (Setae are stiff hairs that in this case apparently increase the effective surface of the paddles.) 

I watched them several times and the following seems clear. There are two pairs of legs that provide propulsion, and those four legs move in pairs of two, so there are almost continuously paddles moving down. The legs have flat surfaces that are held perpendicular to the movement direction on the downstroke and parallel to it when moving up, which no doubt reduces resistance on the upstroke. In other words, the animal is rowing, or paddling, which are both characterised by flat surfaces pushing against water (note that animals such as turtles, penguins and seals do NOT row; their wings/flippers are held at a shallow angle against the water while providing propulsion, meaning they are basically flying through water). 

But what wasn't clear to me was how the limbs open and close; how the paddles are turned to reduce drag on the upstroke; finally, I couldn t really see which legs do what. The camera is jumping about a bit, making it difficult to see what was going on. To have a better look, I selected a part of the video that seemed to have a good quality, and used Matlab to split it into separate frames. I marked the point where the big round tail segment is connected to the rest of the body on some key frames, and defined those points as the centre of new frames. All that remained was to do let the program solve that for 481 new frames, increase the frames in size and make a slow-motion video (10 frames per second instead of the original 29.9). 

 

So here it is! You still have to look at it a few times to understand the movement. I think I see the following:

•  There are two paddle surfaces on each leg. The last segment carries an oval paddle, and the penultimate segment carries an asymmetrical flange. 
• When the folded leg arrives at its highest point, the entire leg is extended and then rotates down in the joint very close to the body. This movement brings the paddle surfaces perpendicular to the downward movement. For the upwards stroke the leg is folded which aligns the paddle surfaces to the upward movement. 
• The two legs that are attached highest up in the image always move in front of the other pair. As the animal is upside down, the legs moving in front of the others are the hindmost pair of swimming legs. We are looking at the belly side of the animal, so these hind legs move more ventrally than the front paddles. In contrast, the front paddles move in a volume that lies more towards the backside of the animal. The volumes in which the four paddle legs move reminded me of Luke Skywalker s X-Wing spacecraft (Bathyopsurus will be slower, though). 
• I was really surprised to see that the front left leg moves in unison with the rear right one. The remaining legs also move together, but at an opposite phase in the movement cycle. I thought at first that the two front legs would move together and that the two hind legs would form the contrasting pair. But no, the legs move in diagonal pairs; if this was a mammal, we would immediately recognise this gait with diagonal pairs as a trot! I can only guess as to why a backwards swimming four-legged paddling isopod use a trot; perhaps this represents a bit of programming, carried over from its past as a walking animal? 

 

Click to enlarge; copyright Gert van Dijk

Click to enlarge; copyright Gert van dijk

Here is a simple scheme of how I think the animal arranges its movement. It is complex! Are there other animals on Earth that move like this? I cannot think of any that swim vertically, and of all animals that swim horizontally, not many use four limbs to propel them. Plesiosaurs and pliosaurs may have used all four limbs, but probably not as simple oars, so the resemblance is weak. On Furaha, kwals come closest, as they also swim vertically and use their limbs as vertical oars. But the three limbs of kwals move together; they do not have Bathyopsurus unique two sets of limbs with alternating strokes. Maybe this means that Furaha needs diploid kwals that do have two sets of three limbs... 

Anyway, Earth has a deep-sea isopod with four rowing legs, arranged like an X-wing, and moving in a trot! I think the three or four hours spent on Matlab programming for this little project were worth it. Why, some may ask? Well, my curiosity was satisfied and I learned something new about animal locomotion. Was this speculative biology? Of course not, it is just regular scientific curiosity fed by lots of speculation.

The 'spietskip', or 'impaler poultry' (Tabulae mortuae VIII, Archives XVIII)

 The ‘spietskip’ is mentioned in the Furaha book; the book contains a short section about the prehistory of the Furaha project, showing old paintings. The word ‘spietskip’ is a made-up Dutch word (the Dutch language allows everyone to make new words by stringing together existing ones). ’Spiets’ means to skewer or impale, and ‘kip’ means ‘chicken’: a ‘spietskip’ is therefore literally a ‘skewer chicken’, although ‘impaler poultry’ sounds more dramatic. 

Click to enlarge; copyright Gert van Dijk

The painting shown above is really old. Over time the animal, with it bird-like plumage and overly reptilian head, fitted less and less within the evolving hexapod scheme. One solution in such cases was extermination, and the other was a redesign. The idea behind the spietskip appealed to me: a small carnivore, hunting from a hiding place among reeds, and catching prey with front legs that are fishing spears. I also thought at the time that The Book needed more scalates. There is ample room for adaptive radiation within every scalate group. Carnivores definitely needed more radiation, to show their range from mouse-sized squigglers to slow massive carrion eaters specialising on dead rusps.

So I had wondered whether redesigning the spietskip might work. An animal clinging to reeds need to have legs with a large movement range, and feet that can grab a reed from about any angle. That should not pose to much of a problem for the general scalate design.

 

Click to enlarge; copyright Gert van Dijk

Here is a simple Zbrush sculpt showing a spietskip following the scalate redesign, including the distal and proximal neck with the neurocranium in between. I now think it is too bulky. It is often difficult to get a good idea of the mass of small mammals and birds, because their fur and feathers make them look much bulkier than they really are. If you see a wet cat, you can see how little mass it really has. Let’s just pretend that the Zbrush sculpt already includes its covering. As you can see, the limbs are very simple; that is because such sculpts were only made to help with perspective and composition, and details can simply be drawn later.

Click to enlarge; copyright Gert van Dijk

 

After importing the body into Vue, I played with limb positions, here simple cylinders, and the overall viewing angle, which was similar to that of the original painting. 

Click to enlarge; copyright Gert van Dijk

 

I could easily have take this image as a rough background and could have started drawing over it, before finally sitting down to paint.

Click to enlarge; copyright Gert van Dijk

However, sometimes drawing without any such 3D aids resulted in livelier images, perhaps at the cost of some perspective mistakes. That’s why I also did a very quick sketch right over the adapted old oil painting. This sketch could also have been the starting place for a new drawing and painting. 

But around that time I realised that I could keep on designing new radiations forever, and that there already enough material for a book already. And that is why the evolution of the spietskip stopped. 

Perhaps I will pick it up again, some day. But first, I am preparing for DinoCon, where I will be selling The Book, images and some other stuff. There are also posts to write on the ever-intriguing theme of how many legs animals should have in speculative biology projects, and on colour changing. 

And for people in the USA, don’t forget that The Book will finally be sold directly in the USA through Simon and Schuster, starting on 28 July, 2026.    
           

Why I dislike AI in speculative biology and palaeoart

This post departs from my usual posts in that it is not directly about speculative biology but about a tool. Let me be clear: I dislike AI and will explain why. Because of the ongoing discussion about AI in palaeoart, I added that subject to the title of the post.

'No AI' icon; you can get it from Wikimedia 

AI as a tool to produce images
Some say that AI allows everyone to paint. I really enjoy painting and the pictorial arts, so wouldn't it be wonderful if everyone could paint and enjoy that as much as I do? But supplying prompts to an AI image generator is not painting; it is merely steering a complex string of associations in a specific direction. Those associations, the core of most LLM models, are derived from countless images, mostly found on the internet. It is no secret that many images used to train the AI models represent copyrighted work and were often used without permission of their creators. With that in mind, you can ask yourselves whether the use of these images is morally and ethically justified. I think not, personally. In fact, you can argue that using the work of others without their permission is a form of handling stolen property.

We have all seen discussions about piracy being a copyright infringement: companies producing films, music, games or software all protested loudly when piracy started to cut into their profits. Their protests gradually made governments fight piracy. But the companies making AI are the ones making money from  piracy, which dramatically affects their point of view. but sometimes I wonder whether governments shouldn't do more to protect copyright, as they did with previous piracy... 
 

Proponents of AI may defend its use by saying that AI does the same as an artist who is influenced by the work of other artists. It is indeed acceptable for an artist to be influenced by others, but there are restrictions: the results are accepted when inspiration leads to creativity. When van Gogh was inspired by Japanese woodblock prints, the resulting oil paintings represented new creativity. In contrast, profiting from the work of others by selling illegal copies of their work is not accepted in art. As AI images are wholly based on associating elements of existing works, they fall in the latter category; there is no inspiration, just a complex way of mixing and copying existing art.          

I will not go into the enormous energy costs of AI (if users had to pay the real costs of producing AI images and videos, hardly anyone would use it); I will also not go into how AI is pushed without any discussion of the moral, social, environmental and economic consequences; but I will say that these aspects make me like it even less. 

DinoCon 2026 AI statement

AI at DinoCon 2026
AI won't be at DinoCon 2026 (I will be, selling images and The Book!). The organisers have made a very clear statement forbidding AI; you can find it here. The image above is what you see when you register as a vendor. I borrowed it from Andy Frazer of 'Dragons of Wales' fame. At the previous DinoCon, Andy brought  'no AI' stickers to DinoCon, and Darren Naish wondered out loud about the ethical and moral aspects of AI use. These anti-AI sentiments have gathered strength over the last year. Good!

AI as a research tool in speculative biology
I have tried ChatGPT a few times as a research tool to examine how useful the results might be. I asked questions that I had already researched the conventional way, meaning I used PubMed and Google Scholar to find relevant peer-reviewed published scientific papers. Two questions I put to ChatGPT were first how deep chromatophores lie in the skin of various colour-changing animals (chameleons mostly), and which kinds of mechanisms might cause quick colour changes in animal skin. For me these are normal things to ask if you take your speculative biology seriously.  

In both cases, what I found led me to conclude that there wasn't much known about these subjects, or at least that I failed to find exact answers in any paper. The AI replies reported more and were much more positive. In fact, ChatGPT congratulated me for asking smart questions, which I now feel is an irritating trick. ChatGPT did produce exact skin depth measurements for chromatophores, and volunteered to put the results in a table or graph 'suitable for publication'. I then asked for specific literature sources, and ChatGPT duly provided five papers. However, not a single one actually existed! I recognised traces of real papers, but only becasue I had already done the hard work. In science, you always have to dig until you find the original measurements that a later conclusion is based on. Well, if you try that with ChatGPT, you get what are mistakenly called 'hallucinations': just nonsense indicating that the association process is going off the rails. If I wouldn't have learned how to find factual scientific information I might have copied ChatGPT's nonsense and would have made a fool of myself. There are other examples of such nonsense in the blog 'Sauropod vertebra picture of the week' (SV-POW; here and here)  

Such experiences suggest that using the word 'intelligence' for what these models do is at best a gross exaggeration. Some AI mistakes are so stupid that no person with any understanding of the subject matter and with a normal ability to reason would ever produce such errors. I now prefer to think that the abbreviation 'AI' stands for 'Associative Incompetence'.

Eighteen years on (also: US availability of The Furaha Book, and Furaha at DinoCon 2026)

 This blog started with the post Why are there humans on Furaha? on 22 April, 2008. That is 18 years ago, suggesting that the blog has now left adolescence and has come of age.

Apparently, Blogger started in 1999, so the oldest blogs could be 27 years old by now. Google bought Blogger in 2003, so the oldest Google blogs can be 23 years old by now. Still, 18 years is a respectable amount of time to maintain a blog, or I like to think so.

The last ‘birthday commemoration’ such as this one was published in 2022. I used to publish a list of most popular posts, but will not do that now, for the simple reason that there are probably so many bots about that the numbers provided by Blogger may not mean much. For what it’s worth, Blogger reports that the blog has been seen 2.63 million times in all, but I have no idea how many bots were among these visits. However, some data are still reliable and meaningful: there are now 329 posts that together evoked 2970 comments. Not bad, I feel, for such a niche subject.

As I read that Google Analytics provides a more reliable view of blog visits and visitors, I activated that 18 days ago. Over that small period, my blog attracted about 1600 views from 522 users per week. Most active users came from the USA (38%), followed, to my surprise, by Singapore with 24%, and the UK with 5%. The blog is in English, so a preponderance of Anglophones is to be expected. I wonder whether that large Singaporean interest is caused by a local interest causing a temporary spike or whether Singaporeans are simply the world largest fans of speculative biology. If any Singaporean readers like to comment, please do!       

Anyway, I aim to continue blogging, even now that The Book is out.

As for that, readers in the USA may have noted that the Furaha book could only be ordered from British sources, and was not directly available from a US distributor. I was told that this was because of contractual reasons, and that the delay would be at least six months. The people at Crowood Press now told me that Simon and Schuster will start selling the book directly in the USA on 28 July, 2026. I am sorry if that is later than you would like; it is certainly later than I would like! 

Finally, I will be at Dinocon on July 25th and 26th, 2026 in Birmingham in the UK, where I plan to sell items there for the first time! DinoCon is of course mostly about dinosaurs, and promises to be even larger and even better than last year!  

 ----------------------------------------------------------

PS (2 May 2026): I thought that Google Analytics would filter out bots. Well, apparently not altogether, as this is what I learned: "Recent trends (late 2025–2026) show a massive, coordinated surge in automated traffic targeting global websites that often appears as direct traffic from Singapore/China".    

Do energy costs affect the size of colour-changing animals? (Shadeshifting II)

This post is not the second post about shadeshifters that I mentioned in the first post on that subject. In case you missed that first one, it was about colour-changing animals; I reserved the word ’shadeshifters’ for animals that can change colour in seconds, minutes, or a few hours at most. The post asked the question whether shadeshifters elsewhere in the universe might be larger than the ones we have on Earth. The promised second post was to be about possible workarounds to allow large shadeshifters. Well, that will -probably- become the third post on this subject.

This new second post will be about the energy costs of colour changing, also called metachrosis. I had mentioned costs aspects without going into detail, for the simple reason that papers on the subject did not provide measurements of the energy bill of colour change.

What would a colour-changing system cost? If such a system on another world is similar to ones on Earth, four cost factors can be distinguished. The first factor is a ‘visual environment analysis centre’; this brain part makes use of an already existing good visual system and analyses environmental visual information in terms of colour, contrast, contours, etc. It then makes a sort of recipe that describes the visual environment. The second part, the ‘body mapper’ is also a brain centre, one that translates the recipe into a map fitting the animal precisely, holding instructions for all colour-producing skin cells (chromatophores). The third factor is the cost of firing neurones that  run from the ‘mapping centre’ to every chromatophore, turning it on or off. The fourth factor deals with the chromatophores themselves: spreading or concentrating pigment granules must also cost something. 

The first three factors have to do with neurons, and on Earth neurons are expensive (for example, the human brain requires some 20% of all blood pumped out by the heart while having a mass of only about 1.5% of the body). I do not know whether neurons on all planets will also be expensive, but let’s assume so. Mind you, if neurons would be really cheap somewhere, an organism would not be punished for throwing brain power at any problem; would that allow a runaway evolution of intelligence? In reverse, are we perhaps hampered by overly expensive neurones? If so, we are perhaps more stupid than the rest of the universe; no wonder they don’t come visiting...

Anyway, back to shadeshifting. Lacking facts, we can still think about whether animal size would affect these costs. Let’s assume that a shadeshifter needs four layers of chromatophores of different colours to produce a wide range of colour effects. I see no need for a large animal to thicken each of these four layers. Whether the colours are visible will depend on the dead tissue above, not on the thickness of the colour-producing layers. A skin area of 10 square centimeter would therefore contain the same amount of chromatophores for a small as for a large animal. In other words, the total number of chromatophores and the number of neurons to turn them on and off are linearly related to the skin area of the animal. If a large animal has three times the skin area of the small one,  it will have three times the number of chromatophores. Note that we just concluded that the third and fourth cost factors linearly depend on skin area. 

Click to enlarge; copyright Gert van Dijk

Aha! Now we are back on familiar ground, meaning ‘scaling’. ‘Scaling’ has been discussed on this blog many times (***). Say we have a chameleon with a length of 25 cm; in other words, L=25. We increase its length, width and height all five times, so the animal become 5 times longer, wider and higher; its length will be 5L=125 cm. Now, surface area scales as a square, so the animal’s skin area becomes larger by the square of 5, meaning it becomes 5^2=25 times larger. The third and fourth cost factors of our larger shadeshifter are therefore 25 times those of the small one.

By itself this does not mean that much. A large animal would in general obviously have higher energy costs than a small one. The key question is how the costs of shadeshifting relate to the animal’s overall energy budget. To answer that, we need to know how metabolism scales with size.

First, we need to realise that each kg of animal will need an amount of energy, so we need to think about the mass of the animal. Mass is related to volume, and volume is easy to work out; it scales with the third power. Our super-chameleon, 5 times the length, width and height of the small one, has 5^3=125 times the volume of the small one, and therefore also about 125 times the mass.

You might think that the energy needed to sustain one kg of animal is the same for a large and a small animal. If that were true, then the larger animal, 125 times the mass of the smaller one, would also have 125 times more energy to spend. That would be very nice for shadeshifting, because the costs of shadeshifting would only become 25 times larger. Brilliant!

But no. It doesn’t work that way.

It actually costs more to sustain one kg of rabbit than one kg of elephant. Small animals have relatively high energy costs, which means that small animals run more quickly out of energy than large ones (this is why mice need to eat often). Mind you, the effect is relative: in total, an elephant of course needs more energy than a mouse; it’s just per kg than the elephants needs less energy. The relationship of metabolism to body mass is well known in zoology:

        Minimal metabolic rate (MMR) = a M^0.75

‘Minimal metabolic rate’ means the animal does nothing besides being alive. The constant ‘a’ differs between groups of animals, and we can forget that for now. The exponent of 0.75 is less than one, which is another way of saying that large animals use relatively little energy. Can we now work out what scaling by length (L) does, rather than by mass (M)? Yes, we can:  mass relates to volume and volume was length to the third power:

        mass (M) ~ volume ~ L^3

We can now replace ‘M’ in the first equation with ‘L^3’, giving:

        Minimal metabolic rate (MMR) = a (L^3)^0.75 = a L^2.25

This is the equation we need. It tells us what happens to MMR if we make the length of an animal 5 times larger. MMR then becomes 5^2.25 times larger, which is 37.4 times larger. Remember that the skin area became 25 times larger. In other words, the energy budget increased 37 times, and the costs for colour changing 25 times. That’s a comfortable margin! In short, large animals can more easily afford shadeshifting than small ones. . 

But how about the first and second cost factors? I see no reason why analysing the visual environment should depend on animal size, so the first factor should not cost more. The second factor was making a map telling each chromatophore what to do. In its simplest form, the costs for such a map would also reflect the number of chromatophores, meaning skin area, which does not alter the reasoning above.

All in all, the relative costs of shadeshifting get smaller as size increases. That is good news for large shadeshifters! Of course, the problems posed by thicker dead skin layers are still there…

 

PS the video at the top of the video was made for the instagram channel I am trying out (j.gertvandijk) 

 

Shadeshifters: colour-changing animals for speculative biology

This video shows an ocopus changing colour;

 I recommend its author's Youtube channel 

Not many animal behaviours on Earth attract more interest than the quick colour changes of octopuses and other cephalopods, who clearly outperforms chameleons and other vertebrates that also change colour. Then again, it depends on what exactly you mean by ‘colour changing animals’. What I mean largely is the ability to use your skin as a high-resolution display screen. That is both attractive and weird, so you would think that many people would use that capability for their speculative biology projects, but that doesn’t seem to be the case (at present anyway). Perhaps the lack of colour-changing is another example of the difficulty of abandoning mammal stereotypes, similar to the automatism of equipping every large animal on any planet with typical mammal front and hind legs? Perhaps...

This post will explore which factors could affect the likelihood of colour changing animals on other worlds. There already is a technical term for this ability: ‘metachrosis’. I admit I wasn’t aware of this term and am probably not the only one: ‘metachrosis’ occurred only four times in PubMed and only 321 times in Google Scholar at the time of writing. ‘Metachrosis’ describes the phenomenon, but we also need a name for animals that do so. The phrase ‘colour-changing animals’ is not precise enough, as we shall see, and also sounds awkward, so I will propose ‘shadeshifters’.

click to enlarge; from Duarte (see reading list)

Not just any kind of colour change

Let’s narrow down the field of colour change in animals. The skin of most humans becomes a bit darker after prolonged exposure to sunlight (in some, short exposure can elicit an unfortunate bright red phase), and human hair tends to turn grey or white if you wait long enough. Many animals change colour twice a year to prepare for summer and winter. Some animals change colour for a few weeks only in the breeding season. The mechanism of these slow changes is usually straightforward in that the amount or nature of pigments in the skin change (not a ‘red’ sunburn though; that is increased blood flow). Such pigment changes are probably universal, i.e. they happen all over the universe, but those are not I mean. I do mean quick changes on a timescale of seconds or minutes, at most hours. I propose to use ‘shadeshifters’ only for animals with this quick type of colour change: cephalopods can change colour within a second, and chameleons take something like 10 to 30 seconds.

Apart from the speed for colour change, the resolution of the display needs to be considered too. In this aspect cephalopods are the undisputed masters, being able to mimic small pebbles, sand, coral, etc. because their skin displays have a very high resolution. But that doesn’t mean that a low-resolution display would be useless or boring. Suppose some animal walks around in a shaded forest without direct sunlight, in which the environment offers little colour contrast as well as little dark-to-light contrast. That animals would do well if it had a slightly mottled almost monochrome colour pattern. But when that animal walks out into bright sunlight, it would benefit from counter shading, with the shaded areas of its body much lighter than the sunlit areas. In both environments the spatial resolution of the skin display could be low, but the change would camouflage the animal well in both surroundings. This suggests that shadeshifting does not need to be limited to high resolution displays to be effective.  

What would you use metachrosis for?

It pays to keep in mind why animals have colours in the first place. Depending on which source you use to answer that question, you may distinguish three or more different purposes. Here is a simple threefold scheme: colours serve thermoregulation, communication or camouflage. Colours for thermoregulation can offer protection from heat, for which being white and reflective helps, or from cold, for which being dark helps. Camouflage is not limited to prey animals that try to escape predation by mimicking the background or breaking up their outlines. Predator can use the same techniques to get closer to their prey too. Animal can also mimic very conspicuous objects to fool others into not recognising them, such as a mantis that looks like a bright flower. As for communication, animals can use colours to signal all kinds of messages to one another, from ‘I am now sexually receptive’, through ‘I am flashy, I am fit!’ to ‘Do NOT mess with me’.

If you think about metachrosis from the point of view of shadeshifters, you start to feel sorry for animals that have no metachrosis, because they must settle for some sad compromise between thermoregulation, communication and camouflage. Think about a snow hare: for thermoregulatory purposes black might be the better colour, but being white in snow offers obvious camouflage benefits (of course, evolution will probably start tinkering with that white pelt to improve its thermoregulatory performance. All this suggests that a major benefit of metachrosis is that an animal can switch between colour purposes. There are indications that chameleons indeed choose different camouflage patterns depending on which predator they face. The mimic octopus uses its colours along with the position of its arms and its behaviour to mimic different animals, and it probably does not choose which animal it mimics at random.

click to enlarge; form wallin (see reading list)

What does it take to change colour quickly?

On Earth, the most common way to achieve metachrosis is to have ‘chromatophores’ with pigment transportation. A chromatophore is a cell that contains particles of pigment. These cells are named for their colours (melanophores for brown to black, xantho- and erythrophores for yellow and red, etc.). Some cells reflect light (irido- and leucophores). When the particles are clumped together, their combined colour effect is weak, but when the particles are spread out, the colour becomes stronger.

click to enlarge; How cephalopods produce colour; from Figon (see reading list)

To choose a particular colour requires turning different cells on or off. Hormones can be used for that but cannot switch a specific yellow cell on while turning the next yellow off. High resolution metachrosis  require a well-developed nervous system than allows ways to control those ‘pixel cells’. This can only be done if the animals has a sort of map, or an instruction tree, to tell which parts should have which colours. That map has to be designed based on the visual characteristics of the environment, and for that you need a high-resolution visual system and appropriate processing power. In short, shadeshifters need a good brain and a good visual system. Don’t expect hi-res metachrosis in animals with poor visual processing capability!

But the rub is that both chromatophores and the nerve endings that control them must be alive. Skins typically consist of layers of living cells that are covered by dead material, such as dead epidermis cells. That dead layer helps provide mechanical protection (and also helps as an osmotic and chemical barrier). It will not surprise anyone that larger animals have thicker skin than small animals, meaning the dead layer is thicker too. That dead epidermis scatters and absorbs light, which ruins the colours of the deeper lying chromatophores. Obviously, metachrosis can only work if the skin is visible! Hair, feathers and thick scales add so much dead tissue that metachrosis becomes pointless. This is probably why the chromatophores of mammals and birds lost the ability to move pigment particles around in the cells: that would be pointless. Fish, amphibians and reptiles do have such cells and do have metachrosis.

Limiting factors

The maximum size of a shadeshifter probably depends on how thick that dead skin layer is. Of course, if colour-changing is very advantageous, you would expect evolution to come up with a way to make those dead cells as transparent as possible, and for the dead skin to be thinner than holds for comparable species of similar size. I tried to find papers comparing the thickness and transparency of the epidermis between chameleons and lizards of comparable size but could not find any. On Earth, large chameleons, about the size of a cat, are the largest terrestrial shadeshifters. It is of course quite possible that chameleon size is limited by reasons that have nothing to do with their colours.    

Animals living in water may not need a thick dead epidermis as much as land animals do, perhaps because there is less mechanical abrasion under water. The largest aquatic shadeshifter appears to be the Giant Pacific Octopus, with a mass of some 15 kg. Again, I have no idea whether octopus size is limited by metachrosis; this does not seem likely, but hi-res metachrosis will have an energy cost.          
           
In conclusion, shadeshifters would be animals with excellent visual processing capability whose skin is open to view, not covered with hairs, feathers or anything like that. Terrestrial ones would probably be fairly small and aquatic ones may be bigger. That size limit is a pity but the remaining restrictions leave  enough leeway to play with. Still, I am thinking of workarounds that would allow larger shadeshifters. Are there ways to make the dead epidermis tougher and more transparent without thickening it? Thinking more out of the box, can the colours from deep in the skin be transported to superficial layers? Farther out of the box, is it possible to change the colour of dead tissue, which probably requires abandoning chromatophores as the engine of colour change? Now that I have already let some cats out of that same box, I would not be surprised if the readers of this blog come up with solutions well before I am ready to write another post on the subject.

Reading list

Wallin M Nature’s palette. How animals, including humans, produce colours. Www. Biosicioence explained 2002

Duarte RC, Flores AAV, Stevens M. 2017 Camouflage through colour change: mechanisms, adaptive value and ecological significance. Phil. Trans. R. Soc. B 372: 20160342. http://dx.doi.org/10.1098/rstb.2016.0342

Figon, Florent and Casas, Jérôme (August 2018) Morphological and Physiological Colour Changes in the Animal Kingdom. In: eLS. John Wiley & Sons, Ltd: Chichester. DOI: 10.1002/9780470015902.a0028065

Stuart-Fox D, Moussali A. Camouflage, communication and thermoregulation: lessons from colour changing organisms. Phil. Trans. R. Soc. B (2009) 364, 463–470 doi:10.1098/rstb.2008.0254

Meyer W, Schmidt J, Kacza J et al. Basic structural and functional characteristics of the epidermal barrier in wild mammals living in different habitats and climates. European Journal of Wildlife Research, 2011, 57 (4), pp.873-885. ?10.1007/s10344-011-0499-9?. ?hal-00664099?

Akat E et al. Comparison of vertebrate skin structure at class level: A review. Anat Rec. 2022;305:3543–3608.