I was inspired by geckos which can sever their own tails.

I wonder if a creature which has such an ability, especially with a high regenerative power to grow its body parts, can sustain itself over and over by only eating its own body parts, rather than as distraction like geckos do.

It may not make it full, or allow it to do many activities, but I hope at least it is enough to keep this creature alive.

  • 13
    $\begingroup$ Ewwww...on several levels. $\endgroup$
    – user535733
    Commented Jul 20, 2019 at 19:39
  • 71
    $\begingroup$ This is the first proposal I've heard for a biological Perpetuum Mobile. The laws of thermodynamics not your fried here. $\endgroup$ Commented Jul 20, 2019 at 19:48
  • 45
    $\begingroup$ Isn't this basically what the body is doing when it burns stored fat? $\endgroup$
    – Adam
    Commented Jul 21, 2019 at 6:55
  • 12
    $\begingroup$ Stephen King wrote a short story Survivor Type for this exact scenario. It isn't good science, but it's a good (and grisly) story. $\endgroup$
    – Graham
    Commented Jul 21, 2019 at 7:20
  • 10
    $\begingroup$ Define "Sustain" or "Survive". Is this a long-term strategy, or a short-term emergency trait (a stop-gap until real food becomes available) $\endgroup$ Commented Jul 22, 2019 at 9:40

12 Answers 12


Sort of. As others have noted, this is essentially what you do when burning body fat.

But you are asking it an organism could INDEFINITELY sustain itself...and the answer to that is no. The laws of thermodynamics prevent it. But, if geckos could eat their own tail, they would have less body mass requiring calories for maintenance, while receiving a temporary boost of calories to store for the future. The body part in question would also need to not have a substantial negative side-effect associated with its removal that negates the benefit of its provided calories: if the organism loses enough blood from removing the limb, even considering the amount of calories gained, it may still result in a total net loss of calories. So the body part needs to be:

  • non-essential to function and/or redundant
  • easily removed without adverse effects (ex: extremely quick-clotting blood)
  • be composed of material that will be readily converted to energy to increase efficiency: fat/bone-marrow > muscle
  • not be regenerated so quickly that the calories gained are too quickly consumed in regeneration of that same part.

So your creature in question could have quick-clotting blood, and detachable body parts that would otherwise serve a purpose as something like decoration for mating-rituals. It could also have an internal system that reacts to generally low caloric intake, and restricts the regeneration of those body parts until caloric intake levels reach an average threshold again. Snakes and hibernating animals would help provide research cases of real-world metabolism regulation.

  • 2
    $\begingroup$ Clever use of body fat as the answer! However, I would note that the OP does indeed appear to be asking for an indefinite or long term solution: it is specified can sustain itself over and over. "Over and over" is as synonymous to indefinitely as we can get. The Rule of Three might apply here: most people can go without food for perhaps 40 to 50 days or three fortnights. Eating yourself to the exclusion of all else is not only a closed loop system but a death spiral. $\endgroup$
    – elemtilas
    Commented Jul 25, 2019 at 17:35
  • 1
    $\begingroup$ @elemtilas Well...I didn't pay enough attention to the question they asked...but I did accidentally answer their question by stating that wasn't possible, and then explained why, while proposing a realistic option to replace the concept... ::shrug:: $\endgroup$ Commented Jul 26, 2019 at 2:41
  • 3
    $\begingroup$ Where is that horrible, horrible question on surviving by eating your hair $\endgroup$
    – Innovine
    Commented Jul 26, 2019 at 8:29


To regenerate requires more energy than the part itself will supply.

To eat the part you have to break it down chemically and then those chemicals are transported through the body and recombined to make new cells.

All of that requires energy.

You started with a severed limb. It cost you energy to digest it and for your body to "build" a new limb from the digested materials.

To get that energy requires you eat something else (or grow limbs that are smaller and smaller each time). Some of what you eat has to be used to process what you eat.

You'll also expend energy severing the limb, healing the wound, preventing and possibly even fighting infection and just staying alive while the very slow process of limb regrowth happens.

This is why, I suspect, regrowing limbs isn't generally used as a survival adaptation in the evolution of creatures. Animals that lose significant parts of their bodies (in nature, without hospitals and medical care) have a higher risk of infection, disease and death not to mention more vulnerability to attack while lacking a complete set of components. It's a generally survival adaption because, mostly, they're less likely to survive, not more likely.

  • 19
    $\begingroup$ aka the law of conservation of energy, aka perpetual motion machines don't exist $\endgroup$ Commented Jul 21, 2019 at 11:11
  • 4
    $\begingroup$ @hidefromkgb NOPE, your argument purely means that there is a way around the first law of thermodynamics. However extracting work from vacuum energy would violate the 2nd law of thermodynamics. $\endgroup$
    – Aron
    Commented Jul 22, 2019 at 2:58
  • 6
    $\begingroup$ @hidefromkgb You mean the same guys who funded people to feed people LSD and try to "see" into Soviet Russia using Extra Sensory Perception? $\endgroup$
    – Aron
    Commented Jul 22, 2019 at 9:50
  • 6
    $\begingroup$ The rule of thumb I was told in school was that you are able to grow 10% of the mass that you consume. Eating limbs might be an expediency but would not be a long term viable solution. $\endgroup$ Commented Jul 22, 2019 at 13:41
  • 9
    $\begingroup$ @hidefromkgb well then you are no longer scientific. Science is based on what we DO know, there's no point arguing about the existence of god or voodoo or perpetual motion machines. - If you wish to argue bring some research which is new and show it is indeed possible; don't waste everyone's time by going pseudo science. $\endgroup$
    – paul23
    Commented Jul 23, 2019 at 3:31

Building on the previous answers, no. But there could be a weird adaptation, where a creature has an "edible" organ/tissue, which grows steadily with the intake of excess food, and can be eaten later to sustain the creature when external food sources are scarce.

We do the same by growing our fat cells. It's more efficient that way, but evolution doesn't always take the most efficient route.

One actual advantage of an externally edible food store would be the ability to feed others (offspring, pack members) with it, or the ability to chew off and drop some of it as a distraction for a chasing predator.

  • 8
    $\begingroup$ On positive note, predator may be even fine with result of getting some easy food this way $\endgroup$ Commented Jul 21, 2019 at 11:47
  • 16
    $\begingroup$ Like milk, in mammalians? $\endgroup$
    – lvella
    Commented Jul 22, 2019 at 13:04
  • 9
    $\begingroup$ Camel humps are largely fat and exist for the purpose of energy storage. $\endgroup$ Commented Jul 22, 2019 at 13:43
  • 3
    $\begingroup$ @ivella lactation is too slow. Sometimes you want to quickly dump nutrients to your offspring and be on your merry way. Like many birds do with half-digested food using a special pouch in their necks. $\endgroup$ Commented Jul 23, 2019 at 13:27
  • 2
    $\begingroup$ Rabbits generally eat their own poop because their digestion is inefficient, and it needs several passes. Cows have different compartments to their stomach, to allow food to go through different processes. And you could even say that an animal eats and digests a food, to grow a external organ, with outside enviroment having an effect on it. Think photosynthesis. $\endgroup$ Commented Jul 24, 2019 at 12:42

The answer is yes and no, but probably not in the way you intend.

The first answer is from thermodynamics: the creature cannot gain more energy from eating the tail than it expended growing it. In fact, due to the inefficiency of metabolism, you will gain quite a lot less.

Then again, nearly every creature alive does this on a regular basis. Mind you we don't engage in coarse measures like gnawing on our own limbs. We can do this much more efficiently with chemical signals. Catabolysis occurs when the body decides to consume its own tissues to survive. This is actually happening on a constant basis with fat cells. Through lipolysis, fat cells are constantly shedding their energy rich fats so that the body may consume it. In more extreme situations, we'll even turn on our own muscles, harvesting them for the protien to survive.

I doubt this sort of chemical reality is what you were looking for. However, consider butterflies and moths. When the caterpillar is born, it has no use for its protective egg shell anymore... but it has valuable nutrients in it. The caterpillar turns around and eats its own egg before going out to chew on the leaves. So this may be an example of a part of the caterpillar, its egg shell, which gets cannibalized because it is no longer needed.

Which leads to one case I can think of where you might actually see a pattern like you describe. There may be a symbiosis of two creatures, like corals and the algae they hold onto. The algae give corals extra energy in the form of sugars, and the corals protect the algae.

However, this balance can shift. When a coral is stressed, such as due to rising ocean temperatures, it "bleaches," expelling all of its symbotic algae into the ocean. This makes it turn white because a large portion of the color of the coral came from the algaes.

You could develop a similar symbiotic relationship. During normal times, the host provides a platform for some energy producer (or other valuable role). This energy producer breeds during times of great plenty. In times of strife, the host may decide that it can no longer provide the nutrients the symbiote needs, and turn on it, consuming it for fuel. As long as the symbiote's lifecycle dovetails with this nicely (I'm thinking something interesting like the jellyfish lifecycle) the symbiote species will not be "offended" by this, and patiently wait for the environment to support the host better.


The short answer is: Yes with magic, no without magic.

Wouldn't you rather have a magical bowl that is always full with fresh nutritious food, perfectly matched to the species that approaches it?

A lot less disgusting and utterly painful for that poor creature. And the magic is the same, or perhaps less.

You need a lot of energy or magic or power to regenerate a body part, and that's in part because it isn't just meat, it's a functional part of the body with nerves and blood vessels and more.

Creating basic food shouldn't be nearly as hard. But it still requires magic or some sort of power source (which can be the sun) to create the food or at least magic or work to harvest and prepare the food to put in the bowl.

It's the same idea. Without magic, you don't get something for free (even with magic there are often limits; it depends on how you set up your magical system).

When geckos sever their tails (for quick getaways from danger) they have to spend a lot of energy to grow them back. Even if they return and eat them, they're not getting anywhere near the calories it takes to regenerate.

Remember, it takes calories to digest food. It's like taking out a loan to pay back previous loans. There's interest on the money you owe and you simply can not sustain the process.

Nor do all the calories you need for regeneration go to edible flesh. There's just too much else you need there. Only a fraction of the energy it takes to grow a body part becomes stored calories in the body part itself.


This happens to an extent when a tadpole metamorphs into a frog.

Most tadpoles are herbivores. They have small teeth that chew plant matter growing in the water. Their gut has enzymes which break down cellulose and other plant tissues.

On the other hand, adult frogs are insectivores. They capture prey with their tongues. Their digestive system has enzymes for breaking down chitin and proteins.

Metamorphosis is much more than growing legs. The frog's entire digestive tract must be reconfigured for the change in diet, from plants to insects. They lose their teeth, their tongues grow, and different digestive enzymes are expressed. So they are unable to eat during this process. (Lungs and rear legs usually develop while the tadpole can still eat.)

Where then do they get the nutrients to survive and finish growing legs during this process? They catabolize the tissues in their tail, absorbing the nutrients. By the end of the process, the tail is gone. An added benefit is there is no tail to interfere with the adult frog's hopping.


It depends on how you define the regenerative power.

If the regenerative power is similar to that of Werewolves, Trolls and that kind of magic regeneration then yes, he could. These types of regeneration rarely take into account Newton's laws and will generate matter and energy on the fly. So if the creature eats its own limbs and regenerates them for "free" then he could easily get his daily food intake.

If the creature has to generate them out of the energy he eats he's going to be dead after the first time he lobs off an appendage. Healing wounds and accelerated growth is one of the most energy intensive things your body can do. That single arm would barely regenerate your arm because of all the energy and material lost on just homeostasis of your body and the digestion and transport throughout your body.


Defying established thermodynamics, and in order to play the Devil's advocate: If you do not assume that "sustain itself" implies "forever", then the answer is yes.

This is something we (and all animals) technically do on a smaller scale whenever we exercise beyond the point where glucose/glycogen is near-fully depleted, and still more energy is needed than beta oxidation alone can provide. We kill muscle cells and feed on the protein.
Needless to say, this doesn't work forever, and it's less than, well, optimal.

On a macroscopic level, it works even less favorably since some parts of our bodies are not easily digestible yet take quite a bit to build. Think fingernails or hair.


Not really. Unless...

Other answers do a fine job of explaining why it wouldn't be possible. Mainly, regrowing a body part costs as much energy as you get from eating it in the very best case (but probably more).

However, if between severing a body part and that body part getting eaten again, energy is added to that body part, it could theoretically work. The creature's metabolism would probably still have to be amazingly efficient (to minimize the losses), the amount of energy added would probably have to be really high (to make up for the losses) and I really don't think any kind of evolution would head in this direction, but nonetheless, it's no longer a physical impossibility.

Such energy could be added in a number of ways. Perhaps the severed body part stores heat or light from the sun. Perhaps it attracts other organisms that are eaten together with the body part. Perhaps the body part absorbs chemicals from its surroundings (maybe a body of water?)

  • 2
    $\begingroup$ Not sure which, but there's a sea creature that everses its entire stomach over the things it eats. Maybe that counts? $\endgroup$ Commented Jul 22, 2019 at 11:59
  • 1
    $\begingroup$ @JohnDvorak Definitely sounds like it's the closest that we can actually see in nature... $\endgroup$
    – Jasper
    Commented Jul 22, 2019 at 12:01

No, but...

If you want a creature that has to eat its own body part to survive, make it the second iteration of eating. It'd be sort of like how cows burp their food back up and chew it a second time, except it burps up its whole belly and leaves it somewhere to digest for a while.


Its a semi-inteligent plant like creature from a alien world with dark and light seasons, possibly a planet with a slow rotation. This makes for nights that last as long as a Earth months, with days equally long.

This creature is kilometers wide, like a forest. It takes solar energy from the sun and grows fruit like appendages. These 5 meter wide "fruits", instead of reproduction, are used to store A LOT of energy for the month of extreme cold and darkness.

The fruits have 1 meter thick skin to protect them, and the real body of the plant is a few meters big, deep undeground, like a carrot or turnip.

The big "forest" part of the plant that is the most visible is also temporary, like leaves. Underground branches connect all parts of the creature. The body and branches secrete a substance around them. This helps with isolation from cold and warm seasons.


Some answers already get rid of the "can you" part. But I guess in worldbuilding, find a workaround is also a good way to answer.

Let's suppose the creature lives in an environment less likely to infection, and that its health is very powerful: it can for example live in desert where the sun is able to kill bacteria in the wound.

Then, let's assume that sun is also used by the creature to process the oxygen it is breathing: sun and oxygen are not limited in the desert, so the creature has a basis on which to grow: Here we are.

To finish, the creature eats a dedicated part of its body regularly, and while the sun prevents any infection, it uses the energy from oxygen and sun to process its own flesh into something valuable for moving and for gaining back its skin. However, the part of its body will not be as consequent as it was before, unless the creature goes back to more sustainable sources of food.

  • 2
    $\begingroup$ Why does it need to eat the body part? That always leads to some loss. Processing internally, as with fat does too, but probably not as much. $\endgroup$ Commented Jul 23, 2019 at 17:09
  • 1
    $\begingroup$ Yes but I was assuming the "eating the body part" was a pre requisite for the OP. What I described is also possible for internal fat-processing $\endgroup$ Commented Jul 27, 2019 at 10:27

You must log in to answer this question.

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