The Hydra is a mythical creature with the ability the regrow severed limbs, including the head(s). According to the dungeon and dragon 3.5 version of the hydra : each time one limb is separated from the body, it grows back. Sometimes, two heads will grow to replace the lost one, until it reaches a certain limit. The only way to kill the beast is either to cut all the heads or to kill the body. It is not possible to kill the hydra just by cutting the same head several times in a row.

Other species can also regenerate limbs, like the salamander. But the regeneration is somewhat different for the hydra since it also regenerate the heads as long as there is still one standing.

So, I have three questions related to the regenerative ability:

  • Is it scientifically possible for the hydra to regenerate her lost heads? I'm asking because it is more complicated to replicate the head than it is for an arm or a leg. This is assuming that they have some kind of brain (in the head) but have the same intelligence as most animals.
  • How long would it take to regenerate one head if it was as big as in the picture above?
  • This is related to the above question: Considering the time and energy required to grow a head, could they really grow several heads in a row?
  • 3
    $\begingroup$ Freaky. $\endgroup$ – HDE 226868 Feb 11 '15 at 1:23
  • 3
    $\begingroup$ Goblin Dan's All-U-Can-Eat Hydra Head BBQ Hut. $\endgroup$ – Schwern Feb 11 '15 at 22:43
  • 1
    $\begingroup$ @Schwern that was my initial idea but it's not possible for the hydra to generate more matter than it consume energy without magic. $\endgroup$ – Vincent Feb 12 '15 at 0:14
  • $\begingroup$ There's a big difference between being able to regenerate heads and limbs in general (which might be explained with science) versus being able to grow lots of new tissue in 1d4 rounds (a matter of seconds). $\endgroup$ – Peteris Feb 12 '15 at 8:27

Is it scientifically possible for the hydra to regenerate her lost heads?

It's simpler if you put the brain somewhere else. Physiologically you'll have to run an optic nerve down the neck, but that's no weirder than having a spinal cord. Psychologically it avoids the question of how a hydra coordinates its thoughts between all the heads. Perhaps the real brain, including memory center, is located in the body and the heads have smaller brains which are only for local control and visual processing.

If you must put full brains in each head, regeneration comes with the additional problem of not just memory loss, but having to retrain each brain up from infancy to learn motor control. You can make the motor control recovery time shorter by making hydra intelligence and adaptability low, they're simply born with brains as developed as they need to be. Memory loss can be mitigated by the Hydra treating their brains like a disk RAID with each memory being stored in several brains.

Either way, regrowing a limb has to deal with the problem that the limb must remain protected from the environment while it is growing, that's a fancy way of saying it needs skin. As the neck and head grow, it would probably have to shed its now too tight skin many times like a snake.

How long would it take to regenerate one head if it was as big as in the picture above?

The Internet consensus is it takes a gecko about two months to regrow their tail. Larger animals tend to grow slower than smaller ones, your hydra is a lot bigger than a gecko, and the cube-square law is solidly against you on this one. As an animal doubles in size its volume (and thus weight) increases by something like nine times. So a lot longer than two months.

Using the humans for scale, I'm going to guess that head is about 2 meters high and weighs about the same as a horse, or about 400 kg. It takes about two years for a horse to mature so I'm going to say in the order of a few years for a hydra to regrow a head.

The neck probably adds another 6 meters and maybe 1200 kg which will take even more time to regenerate. If you want to make sure that hydra doesn't bother you for a good long while, aim low.

Considering the time and energy required to grow a head, could they really grow several heads in a row?

Yes, if they ate enough. It would be a strain on their body, but I don't see how it would be much different from the multiple pregnancies, or the normal process of growing. All that bone and teeth will be a big strain on calcium, so they'll need a lot of that.

1 kg of beef contains about 2300 calories. A 400kg head would contain about a million calories plus essential (ie. the ones the Hydra cannot make itself) vitamins and proteins plus minerals. It would need to eat likely more than ten times that to regrow it and maintain itself while doing it. If you really want to figure it out, look into how much you need to feed a large, growing carnivorous reptile.

  • 6
    $\begingroup$ I love the idea of raid brains! $\endgroup$ – Liath Feb 11 '15 at 13:41
  • $\begingroup$ @paqogomez It doesn't, it dies. If you cut off all the heads, and they contain its brains, it will die. If the brain is somewhere else, without a really big IV it will eventually die of thirst; I don't believe the hydra has enough internal reserves to regrow a head and sustain itself for the year(s) it would take. $\endgroup$ – Schwern Feb 11 '15 at 20:34
  • $\begingroup$ @paqogomez The lore behind a hydra states that you kill it by cutting off all of the heads. That fits in with this description fairly well. $\endgroup$ – Thebluefish Feb 12 '15 at 0:16
  • $\begingroup$ Growing tissue takes a lot more energy than what the tissue contains. Even for our current meat farming animal species, which are optimized to grow lots of meat quickly and efficiently, you need to spend 10-50 calories of food per each calory of produced meat, so instead of "plus some percentage" a hydra would need to eat (and hunt!) multiple tons of food for each head replacement. Eating a party of adventurers and their pack animals would be just a small part of replacing a single head. $\endgroup$ – Peteris Feb 12 '15 at 8:36
  • $\begingroup$ @Peteris +2000% is a percentage. :) I agree, but I don't know enough about the metabolism of reptiles to know what that number is. 10x is probably a good start. I've updated the answer to make it clearer. $\endgroup$ – Schwern Feb 12 '15 at 17:35

Scientifically it is not possible to regrow limbs in the sort of time-frames we are talking about here, however the same effect could be achieved at least for a while.

Have an alien creature with the main brain inside the body. Each head only has the intelligence required to improve its mouth-eye co-ordination for reflexes. The necks do not use bones but instead have a hydraulic reservoir that is pumped full of liquid to inflate them and fill the role of bones.

The hydra actually grows many heads and stores them coiled up and deflated inside its body. Due to space and energy constraints it normally only has a few heads active at any one time.

A natural reflex should one head be severed then immediately has it push out and inflate two more heads from its internal storage. To someone watching the heads would appear to emerge and grow extremely rapidly and replace the one head with two.

Once the threat is over it would deflate and return some of the heads to storage and gradually regrow the missing ones and replace any lost hydraulic fluid.

Why would something like this evolve though? I suggest two mechanisms that reinforce each other:

  1. Food is hard to find, and a lot of creatures burrow. They use the heads to send them down burrows to haul the creatures inside out. This does mean the heads risk getting trapped or damaged though so they evolved to be able to abandon trapped heads (ideally pulling them out and eating them to reclaim the nutrients) when needed.

  2. Many heads meant you were a successful hunter. It became a large part of their mating displays, as a result they evolved to grow more and more heads - but that wasn't sustainable all the time so they worked out how to collapse them when not in use. Eventually that evolved into the storage system.


Is it scientifically possible for the hydra to regenerate her lost heads?

Yes, however, many things have sound science, but the practice of it is usually what holds it back, which the other 2 answers will explain.

How long would it take to regenerate one head if it was as big as in the picture above?

This would depend heavily on the last question. Sorry if this answer seems to be dodging the question, as this answer is heavily dependent on energy

Considering the time and energy required to grow a head, could they really grow several heads in a row?

Imagine the human body, it can repair minor injuries, but this requires nutrition from food, and takes quite some time. So, now the hydra needs 2 things, mass and time/energy.


Scaling this up to an entire head of beast of that size, they would at least need mass equal to their lost heads. Note the at least, it is not enough to have 500 kg of protein in your belly to regenerate the head, as the head contains noticeable other things, such as bones, teeth, the brain etc.

So, the Hydra would need to be extremely bloated/fat, containing enough nutrition within it to regenerate it's heads. However, this is great, for your question, as that means that there is an actual scientific end to their head regeneration. If they run out of nutrition, they can not regenerate anymore.


Alright then, let us assume the Hydra is the product of billions of years of evolution, or genetic engineering, or something to that effect. Because of that, the hydra simply needs to provide calories to be able to move the nutrition from her belly to her head and not worry too much about minor things like biology.

Well, you know some of the simple physics equations, Force is Mass * Acceleration, Work is Force * Distance, Power is Work / Time. With those, you can find that depending on how much calories the hydra focuses into the head regeneration, the time taken to regenerate also varies.

Some quick and not too specifically accurate high school level science:

  1. Suppose that the head weighs 100 kg. (It should really be much heavier, but this is simpler)
  2. Suppose that the distance from the belly to the head is 5 m. (Do note that this example hydra is much smaller than the one you want, but you should be able to scale it up)
  3. To regrow the head in 1 s, the amount of force is 100 kg * 5 m/s-2 * 2 = 1000 N. The 2 is to stop the mass at where the head is.
  4. The work done is 1000 N * 5 m = 5000 J
  5. The energy required is 5000 J / 1s = 5000 W.

So, the amount of work needed is not too much, only about 1000 calories, but the power output is quite sizable.

However, in the end, the power a creature can generate is largely dependent on it's food, so the limiting thing for your hydra would be how fat you want it to be. Either that, or give it super metabolism and have it eat things with some of it's heads while fighting to fuel it's regeneration


It's certainly possible to regenerate. What's not possible (sans magic) is for them to regenerate quickly. In addition to the nutrition aspects mentioned by others, there's also the matter of thermodynamics. Healing/regeneration is fundamentally a bunch of chemical processes, and those processes produce waste heat. (Even notice how the area around a healing cut will be warmer than the rest of the body?) Do that too fast, and the creature cooks what it's trying to regenerate.

Brains wouldn't necessarily be a problem, if we assume that they've evolved a distributed parallel system, like a RAID disk storage system. All memories &c are mirrored across the brains (perhaps with a bit of time lag), so the regenerated brain's content is restored from the existing ones. Note, though, that cutting off a significant number of heads would seriously degrade the total, and could result in partial amnesia, depending on just how much redundancy there is.


The only way I can see that you'd get a hydra that could regenerate lost heads in a timeframe sufficient to employ the regenerated head against the enemy that chopped it off is actually quite simple:

Do not permanently bind the creature's cells together.

In conventional organisms, a creature's cells are bound to adjacent cells in a fashion that is difficult to reverse. This makes the bond strong, but slows healing.

Now, if we have cells that can bind to one-another reversibly, we have a whole different organism. That sword-swinging pest chops off one of our heads. The head, recognising that it is the lesser of the two pieces, proceeds to disengage its intra-cell bindings and reform so that it can rejoin the main bulk of its body as soon as possible, or failing that, become a smaller organism. The main organism could re-assign cells to reform to regain a lost organ or limb, becoming unimpaired but smaller due to the lost mass - until the lost mass rejoins. These cells would have to be a little more complex than our own, but not greatly so.

This would also account for the mythological method of dealing with the regeneration: burn the stumps, and eventually the whole creature. Cauterising the wound would kill a lot of the cells and slow down the business of regeneration. However, this mightstill not account for what would happen with the severed heads, which would have to be dealt with separately.

Why would this creature have a specific body plan and not be an amorphous blob that could do anything? The answer is the simple fact that the cells are bound together. It would take energy to unbind them, which could be better used for other things most of the time. The time taken to regenerate after an injury would depend on the magnitude of the injury, but small cuts might regenerate in a few seconds, and a major wound might regenerate in minutes, probably less than an hour.

Depending on the method used to repair injuries, and the magnitude of the injuries, it may not take much energy at all, or it may involve the complete reconfiguration of the body. Re-attaching missing extremities would be relatively cheap in terms of energy, while extensive body remodelling would be more expensive. A clean cut actually affects very little volume of cells, whereas a burn affects far more.

The main flaw of such a creature would that they would not be quite as robust as a permanently-bound creature. Depending on the strength of the intracellular bonds, it may be possible for a permanently-bound organism to physically tear a reversibly-bound organism apart rather than simply cutting it, however this would be as readily healed as any other injury.


An article recently posted to Cracked, "5 Fantasy Beasts That Wouldn't Work (According to Science)" by Justin Crockett and Nathan Murphy, points out that a hydra might collapse under the weight of its heads:

The hydra can have as many heads as it wants, but it still only has one body, and it can only support so much weight. You try sprouting 20 more heads and then fighting a muscle-bound dude with a sword.

This means you'll have to make the heads light and not overly slow to regenerate. That got me thinking: what could look like a head but not actually be a head? (Consider the hind flippers of a seal: they perform the function of a fish's tail but are instead homologous to its ventral fins.) You could make the "heads" out of arms.

  • Limbs would regenerate if amputated, as you pointed out for a salamander. Regeneration without magic is slow, but stories can be paced with other reasons why predatory pests (such as your heroes) might only occasionally being able to score an amputation.
  • Limbs would have enough nerves to process reflexes. A newly generated limb would need some rest time to build up "muscle memory".
  • Fangs would be analogous to teeth but adapted from claws, grinding things before feeding them to the smaller true head.
  • The features analogous to eyes would be adapted from other claws. There are real skin conditions that cause skin to be light-sensitive. If a claw is adapted into a primitive lens over photoreceptor cells, reflexes could make the hand behave as a head that can "see" a predator.
  • Cutting off the true head will kill it. But that's far easier said than done, especially with all those arms protecting the head. So "having to cut off all the heads" is more of a practical matter of cutting off all the arms before the head can be reached.

The head/arm analogy brings up interesting concepts for how it fights off random predators carrying a sword. One "head" could grab the pest and hold it, and then another "head" grabs the other end and rips it apart.

  • 1
    $\begingroup$ Rather than arms, they could be analogous to an elephant's trunk. Perhaps something like the star-nosed mole: en.wikipedia.org/wiki/Star-nosed_mole Perhaps a number of 'heads' retract into the body, and are extruded as needed... $\endgroup$ – jamesqf Feb 11 '15 at 21:23

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.