This question considers the respiratory apparatus of an aquatic (or amphibious) species descended from engineered humans.

The history of the species provides a relevant constraint. While the science in the setting is soft and the handwavery of fictional genetic engineering might be called upon to produce nigh arbitrary results, I would much prefer, if possible, to be able to claim “these structures that serve [some function] are actually homologous to [some other structure found in humans]!”, or “similar solutions are found in [some other species].”

Given these conditions, what would be the reasonable changes to make to humans such that they would be able to breathe underwater?

Relevant information come across during the search for an answer:

I recall seeing objections to the practicality of gills (on the neck/beneath the jawline) citing the square-cube law, even though I am unable to locate the source text. I am also uncertain about the specifics of this claim (might it still be possible for, say, large external gills to provide enough surface area?) — I’d be grateful if anyone could verify this or point me to a source. Another issue with modified humans with authentic gills is that they would likely have to lack ears, tonsils, and the thymus, since those are developed from the same structure that give rise to the branchial skeleton in fish, and it is unclear what impacts this would have on the rest of their physiology.

Some of the arguments presented above are also applicable to the option of having both gills and lungs.

Gills placed instead on the torso (to allow space for larger gills) is an often seen variation; and structures such as “gillungs”, as proposed in this answer, seem to be regarded as feasible. However, these appear difficult to justify from a developmental point of view (to have lungs connected to openings on the side of the torso involves a change in topology, and there would likely be complications associated with punching through the chest cavity?)

Finally, please note that this problem is not about how whale-humans are likely to be more practical than merfolk with gills: air-breathing aquatic humans exist separately in the same setting.

  • $\begingroup$ I might be wrong, but it looks like fish gills are above the ribs. theonlinefisherman.com/images/stories/Images_Stories/… Combine this with the fact that mammalian lungs are well protected by the rib cage, and the fact that -- as others have mentioned -- we need a lot of O2 which the ocean doesn't have enough of, and it's pretty obvious that what you ask for is not anatomically reasonable. So, just have them in your story. $\endgroup$ – RonJohn Oct 26 '17 at 3:20

So, it's really hard to breathe water. The biggest problem is not where the gills are on your body but instead how gills might exist at all. The fundamental limit you'll run into when designing something like this, whether it's through biological engineering or something like artificial gills is that water only has so much oxygen in it- around 10 mg/L. Humans require a LOT more than this- mainly because we're not polikiothermic and use our metabolic heat to stay warm. We run through 10mg/L very rapidly, so we're going to have to keep oxygenated water flowing over our gills to survive.

Essentially, there are three ways to solve this problem:

1. Make the humans really efficient metabolizers.

You could reduce the human dependency on oxygen in two ways- one, by engineering a more efficient method of anaerobic metabolism, or two, by getting more energy from the small amount of oxygen you're getting from the water. Both of these are conceivable, but would be very hard- there's a powerful evolutionary advantage to be gained by getting the maximum energy out of your food, so nature's already optimized a lot in that avenue.

2. Increase the respiratory area

Give. The. Mermaids. Wings! Rather than trying to push huge amounts of water through the mermaid body like fish do, make the gills external. As they flap their wings, gas exchange happens incredibly quickly through the thin membranes in their wings and they can extract oxygen from the water that way. This is a fairly common method for ocean organisms, especially the lovely nudibranchs, which are literally named after their external gills- nudi- naked, branch-lung. They're also beautiful:

Nudibranch, from Wikipedia

Those plumes on their back are their gills- no water filtering necessary! For mermaids, I'd imagine that their wings are thinner and flatter, more like the classic "faerie" wings. They wouldn't necessarily be used for swimming because they'd be quite delicate, but it's certainly possible to reinforce them with musculature that allows them to be useful for mobility as well.

If you're hesitant about giving mermaids wings, you could also reuse the thin membranes already in place on the mermaid's tail. There, they'd have a similar mechanism and it might actually work better while swimming rapidly. However, they'd have to keep flicking their tails back and forth even when standing still, whereas with wings they could gently flick them back and forth while staying suspended in the water column.

3. Remove their dependency on oxygen altogether

Not everything in the ocean needs oxygen. You could have the mermaids form symbiotic relationships with zooxanthellae (algae), like corals do, or have them endosymbiotically ingest chloroplasts- they'd have green skin but that might be totally okay. That would also limit them to the sunlight region of the ocean, however, and that might be a bigger problem.

If you're prepared to do some more intense genetic engineering, you could make them entirely chemosynthetic- using elements like sulfur to obtain metabolic energy. They'd need to eat a lot more sulfur (as much oxygen as we consume daily) but for an advanced civilization that wouldn't be a problem. The general equation for chemosynthesis is $H_2S + CO_2$ -> $C_6H_{12}O_2 + H_2O+ S_{(s)}$.

  • 2
    $\begingroup$ I just want to point out the absurdity of the 3rd option. Its possible with GE however redesigning a complex organism like a human and changing their metabolic chemistry is a HUGE far reaching alteration that stretches thin the line of reasonability. $\endgroup$ – anon Oct 26 '17 at 15:20
  • 2
    $\begingroup$ @anon genetic engineering always sounds like magic. In this case, it might not be as much genetic engineering as applications of endosymbiosis theory. That’s how we got mitochondria and plants got chloroplasts, and I don’t see a reason that’d be impossible for an advanced civilization. It’d be complicated by the multicellularity of humans, but there are ways around that too- inject the zygote, maybe, or bathe newborns in some kind of chloroplast bath. $\endgroup$ – Dubukay Oct 26 '17 at 17:54
  • 1
    $\begingroup$ @anon the chemosynthetic idea is probably the more plausible of the two. All you’d need to do is find a bacteria/archaea that metabolizes sulfur and add it to the gut microbiome. Then, the newcomer would be passed to offspring via mother’s milk or one-time injection. This has a few archetypes in nature already, such as the deep-sea tubeworms on mid-ocean ridges $\endgroup$ – Dubukay Oct 26 '17 at 18:00
  • 1
    $\begingroup$ I never said it was unrealistic, I questioned its reasonability. You can easily quantify reasonability in GE with the amount of changes/biological steps it would take to reach the end result. Messing with the fundamental metabolic structure would likely be one of the most difficult and massive projects. $\endgroup$ – anon Oct 26 '17 at 19:17
  • 2
    $\begingroup$ So, probably too late to add this and inspire a full discussion but as a followup thought. Adding wings to mermaids will increase their respiratory capabilities, but will also GREATLY facilitate heat loss, which is the whole reason we needed the oxygen in the first place. I don't know the thermodynamics of the whole system and it's possible it'll still work but it isn't quite the silver bullet I initially thought it was. Hmm. $\endgroup$ – Dubukay Oct 30 '17 at 20:56

Use the lungs.

from http://archive.rubicon-foundation.org/xmlui/bitstream/handle/123456789/5903/SPUMS_V5N3_2.pdf?sequence=1

The entire chamber was pressurised with air and an anaesthetized dog was lowered into a tub of oxygenated saline. The animal was kept cool at about 32°C in order to reduce his oxygen requirement. While submerged, the dog continued to breathe, and jets of water rising from the surface showed clearly that he was pumping the solution in and out of his lungs. At the end of the observations, the dog was lifted out of the tub and his lungs were drained of water and re-inflated with air. One of these dogs was later adopted as a mascot by the crew of the Royal Netherlands Navy vessel HMS Cerberus.

This was some cool reading. I knew that animals (and humans) could breathe oxygenated perfluorinated liquids indefinitely. I did not know that before that there was research done with water. Issues with breathing water.

1: If osmotically much different from blood it damages the lung tissues. So these experimental animals (and people!) breathed saline solutions isotonic to blood. OK.

2: It is hard to get enough oxygen into the saline. They had to use high pressure hyperbaric chambers. I read an estimate of 160 atm to get enough O2 into saline to meet O2 requirements for a mammal. That is about 1 mile deep.

3: Surprisingly (to me) it was the gradual build up of carbon dioxide that killed the experimental animals. CO2 is not very soluble in water. We can blow off our CO2 by hyperventilating.

Obviously then if we put all these factors together, we find that in order to maintain his arterial carbon dioxide partial pressure within tolerable levels - to prevent a sense of suffocation or even loss of consciousness - a water- breathing diver would have to move a substantially greater volume of water per minute in and out of his lungs than the air-breathing diver moves air.

Apparently that does not work underwater. As an animal breathes harder, it reaches the limits of what the pliable airway structures can withstand. They collapse in (sort of like asthmatics have their airways collapse in when they try very hard to inhale - thus the wheezing) and so limit the water that can be moved back and forth.

The merfolk will breathe like people breathe, moving water in and out of the lung.

  • These merfolk may be confined to depths where pressure is great enough that the dissolved O2 will meet their needs. Sort of like air breathing humans are confined to low altitudes, for the same reason.
  • The merfolk have saltier blood - isotonic to the seawater they are in.

    • The merfolks have big respiratory muscles - good for forcefully expanding the chest as well as forcefully contracting it - this latter piece is not especially strong for humans who are much better at inhaling.

    • Merfolk have comparatively rigid bronchi which withstand the higher pressures associated with more rapid inhalation and exhalation of water.

  • 4
    $\begingroup$ "The merfolk will breathe like people breathe, moving water in and out of the lung." That would be really slow. I mean REALLY slow. (There's a reason that fish gills vent out the sides.) $\endgroup$ – RonJohn Oct 26 '17 at 3:07
  • 2
    $\begingroup$ haven them excrete slime that contains the the excess co2. Since the sweat glands aren't needed anymore they could have been repurposed to excrete a mucus that gets rid of the excess co2. $\endgroup$ – Tschallacka Oct 26 '17 at 11:39
  • 3
    $\begingroup$ The problem is that the "pressure" in the experiments was pushing extra oxygen into the water. In the ocean, the pressure goes sky-high - but the partial pressure of of oxygen falls as you go down :-( $\endgroup$ – Martin Bonner supports Monica Oct 26 '17 at 12:42
  • $\begingroup$ @Martin Bonner - are you sure? It looks like it falls but then rises again in the deeper ocean. en.wikipedia.org/wiki/Oxygen_minimum_zone#/media/… $\endgroup$ – Willk Oct 26 '17 at 23:22
  • 1
    $\begingroup$ @NoName - that might have been true 500 million years ago but no longer. The oceans have gotten saltier since our ancestors developed closed circulatory systems. Ocean is about 3.5% salt and our blood is only 0.9%. If it were the same, seawater would be like Gatorade to us. $\endgroup$ – Willk May 31 '19 at 1:36

Lots of nice answers here! One problem which was pointed out is that humans, being warm-blooded, require much more oxygen than polikiotherms, because maintaining body temp requires a lot of metabolism, which in turn burns a lot of O2. This results in all kinds of special strategies to get enough "gillage". So...

Why not make them polikiotherms (cold-blooded, for our purposes), like almost all fish? This reduces a lot of problems. First, less oxygen usage, meaning less extreme solutions for gills. Look at your friendly reef shark:

gills gills gills

Its gills are pretty reasonable in size.

Second, being cold blooded means you don't need all that insulation. Most marine mammals have extensive blubber deposits to help them keep warm, and this doesn't really fit the mermaid image. Mermaids are expected to have a touch of avoirdupois, but tradition has it more fetchingly distributed and as much ... decorative as utile.

Third, cold-blooded mermaids have less need for food, so they can spend less time foraging, and more time perched on rocks luring sailors to their doom.

  • 3
    $\begingroup$ Not all fish are strict polikiotherms. Tuna, for instance, maintain a core body temperature between 25 and 33 C. They're not true endotherms because they don't maintain one constant temperature, but they maintain a relatively high temperature even in cold water. Great whites and makos are homeotherms due to their circulatory system. $\endgroup$ – Keith Morrison Oct 26 '17 at 5:16
  • $\begingroup$ @KeithMorrison yah you got me on that one, but c'mon ... almost all fish are cold-blooded, nice discussion here: quora.com/Are-all-fish-cold-blooded $\endgroup$ – akaioi Oct 26 '17 at 5:26
  • 2
    $\begingroup$ I looked, but was completely unable to find a friendly reef shark. The doctors and nurses were all real friendly though... $\endgroup$ – T.E.D. Oct 26 '17 at 12:10
  • 1
    $\begingroup$ @T.E.D. science never fails to deliver: telegraph.co.uk/news/newstopics/howaboutthat/5428522/… . $\endgroup$ – akaioi Oct 26 '17 at 15:03

The main problem you'll run into is that the human brain takes a lot of oxygen to run, and there isn't nearly as much oxygen in water as there is in air. So if these merfolk are to be able to breathe underwater without surfacing, they'll either need much more surface area on their gills than humans have in their lungs, or they'll need to pump a whole lot of water through their gills.

According to the "gillung" answer you linked, freshwater has about 8 cm3/L of dissolved oxygen, while air has 210 cm3/L. That's a factor of 26.25 difference. So if you want merfolk with internal gills replacing their human lungs, they're going to need to pump over 26 times as much water through their lungs as humans pump air. That's going to be all kinds of impractical, even when they're not exercising. Adding slits between the ribs to expel water to keep it flowing one way through the gills will help, but they're still going to need to pump a huge amount of water in order to stay conscious. And pumping water like that will take a lot of energy, increasing demand for oxygen.

Solution: External gills, and lots of them. According to that Wikipedia page, external gills on the creatures that have them (salamander, lungfish, and bichir larvae) typically take the form of 3-4 fernlike branching stalks coming out from either side of the creature's body, behind the head. I'm not exactly certain where that would correspond to on a human- maybe on the neck? You could probably get away with putting them on your merfolks' backs just as easily, which has the advantage of being closer to the pulmonary arteries/veins that they'll need to hook into.

A system of external gills capable of keeping up with a pair of human lungs is going to have to be pretty huge, so these guys aren't going to be hydrodynamic. They'll also need to move around regularly in order to keep from using up all the oxygen in their immediate vicinity, and since they're genetically engineered from humans, may not have the instincts to do that subconsciously. So it might be better to have them live in rivers, so they won't have to worry about it. Alternatively, they could be engineered to be able to smell oxygen with their fluffy tentacle-like gills, which would be pretty cool in its own right. That's up to how far you want to take your genetic handwavery.

Another alternative could be to do away with the forest of external gills and give them either fluid-flexible crablike gillungs or smaller external gills, and make them able to survive on the small amounts of oxygen that their gills would be able to provide. When they're resting or need to dive deep for whatever reason, they could go into a low-power mode, conserving oxygen and relying on their gills; but when they need to hunt, fight, or do math problems, they could surface for a breath of oxygen.


Fish pull water in through their mouths and push it out over their gills in order for the gills to be able extract the oxygen (or something to that effect). This would not work for humans since the amount of water they could pull in would be limited by the size of their mouths and possibly noses.

I think the external gills ideas floated by @WilliamKumler and SomeoneElse37 would be the most practical solution.

While wings would probably be aesthetically pleasing to look at (I do love the image that brings up in my mind), they might be less practical due to a smaller attached area versus their length and width (thinking of fairy/dragonfly wings here. @WilliamKimler, you might be picturing it differently).

Gill tendrils (like those on the gorgeous nudibranchs), gill feathers (as sported on newts and salamanders) would be be the way to go, and considering that your engineered merperson would likely need to be in an area with some amount of water current, the tendrils can passively sway in said current, or even when the merperson is moving, but have limited motion in the event that the merperson is stationary in an area with limited water movement (like a cave, for example).

Along similar lines, you can cover your merperson in fine hairlike tendrils, covering a larger area of their body surface. Or the tendrils can be engineered in some way to be derived from our human hair and can therefore be situated (mostly) on the head and tie in to the existing breathing system in some way.

The limitation of the external gills would be that they would have to be kept wet, or perhaps be retractable in some way. Alternatively they could go to sleep or become inert when the merperson finds them-self on dry land or exude a slime or gel of some sort to protect themselves from drying out completely.

Finally, speaking of noses, ears and other protruding bits and pieces, you would probably want to make those smaller or in some way bring then in closer to the body to improve streamlining. If you don't want to go with a mermaid-like tail, your merperson would need larger, finned hands with longer fingers, and flatter/larger feet. They might also need to be more muscular, as propelling yourself through water is harder to do than propelling yourself through air.


Or, you could simply do it like whales, which do breathe air, but have the ability to store vast amounts of oxygen internally, for extended stays under water. I believe the sperm whale has been known to submerge for over an hour, when they go to great depths to search for squid.

A technique used by seals in arctic areas is to breathe out the air on the underside of the icepack in a big bubble, let the water refresh the O2 content, and breathe it back in. Perhaps a merperson could have a skin flap they could exhale and inhale a bubble of air into, for this purpose.

With some air breathing aquatic reptiles, like turtles or alligators, their lower metabolism means they don't need to have as much oxygen on hand. But, that's a radical biological change, whereas adapting the merperson to be able to store far more O2 from air breathing would be a more likely change... as it has already happened when mammals returned to the oceans.

Why would you want the merperson to breathe air instead of gills and water? Air transfers O2 far more efficiently, giving the creature greater strength reserves underwater. The largest non air breathing aquatic creature is the whale shark, at around 40 feet, and it is a very slow moving creature with a somewhat narrow range of habitats. Contrast that with the blue whale, about twice the size, that can move faster and venture into a much wider range of habitats, like the antarctic.

  • 1
    $\begingroup$ Do you have a cite for seals under ice? My father was an antarctic seal biologist, and wrote a book on the subject - I'm pretty sure I would have remembered that. (But it may only have been discovered recently). $\endgroup$ – Martin Bonner supports Monica Oct 26 '17 at 12:46
  • 1
    $\begingroup$ If they do do it, I suspect they are letting the water remove the CO2, rather than replenish the O2. $\endgroup$ – Martin Bonner supports Monica Oct 26 '17 at 12:47
  • $\begingroup$ ... and hey! There's no reason the OP's merfolk can't do it, even if real seals don't $\endgroup$ – Martin Bonner supports Monica Oct 26 '17 at 12:51

I would go with PEM electrolysis

A gill like structure with the body ability to produce electrical current can be used to derive oxygen directly from the water molecules (H2O -> H + O2). This structure can pump the resulted gas directly in to lungs (even fairly human like lungs) and can be shunted of when outside water and exposed to free oxygen in the air.

This has the benefit of not being dependent on the level of oxygen dissolved in the water and therefor allows survival in oxygen poor water like high depth or contaminates.

This is a relatively energy inefficient process but is conceivable given the complex structures living cells can form even mimicking amphibian skin respiration (absorbing oxygen from the water via the skin).

Also the existence of such organism (humanoid this case) can support symbiotic hydrogen based ecology with a possible micro re-claimers (H+O2 -> H2O) in areas with natural high oxygen content in the water (presumably near the surface).

  • 1
    $\begingroup$ Hydrolysis, huh? That's an interesting proposal... though its feasibility depends on how much electric voltage/amperage is needed to cause the hydrolysis, and whether any animal could generate that much electric power without harming/killing itself. $\endgroup$ – MarqFJA87 Aug 24 '18 at 16:32

Gills on the back would be my go to.

Creating a structure that traverses or even changes the human chest cavity isn't unreasonable. There are currently gaps where muscle and flesh exist between the ribs. To simply say an oxygen exchanging membrane traverses the back ribs isn't to farfetched. Or simply cavities through the back ribs that bring water to the gill/lungs.

The added advantage in simplicity is that because its located closer to the lungs you could reuse a lot of that organs structure. You could even likely keep both breathing methods, there are afterall creatures who can breath both air and water.

As for other traits. just because you add gills does not necessitate the loss of ears and other organelles. Did you know whales still have their hind legs.

As for making this more reasonable. I would increase the space between the ribs on the back (so a larger more pronounced chest) to allow a bigger gill slits. I would grow the lungs to be large and thin lining the back of the internal chest cavity, so you would have more functional respiratory surface area. The shape of the lungs is already conducive having 2 distinct halves with separation in the middle which would be needed to connect the central nervous system to the rest of the organs.

If going with the cavity ribs notion then your only challenge is ensuring water circulation. Fortunately, we already have an organ that solves this called the diaphragm. Just modify the diaphragm to be drainable/floodable so it can operate in chosen environment.

  • $\begingroup$ Whales indeed retain limb rudiments. However, the suggestion that gilled merfolk would lack ears results not from the assertion that they would have no use for the organs, but from the fact that the structures from which they derive would, in this case, have developed instead into gills. $\endgroup$ – Lok Oct 25 '17 at 22:12
  • $\begingroup$ Im not sure I follow the logic of that one, If we are talking about genetically engineering humans then the only limitations are what is biologically plausible. Reasonability can be quantified in how few changes you could make before you reach your desired effect. $\endgroup$ – anon Oct 25 '17 at 22:17
  • 1
    $\begingroup$ I think I should say that I'm not sufficiently acquainted with the relevant concepts in biology that deal with this. But in exactly that way of quantifying how reasonable things are, I was under the impression that "tell these bits to develop into gills (like they do in fish) instead of ears (like they do in humans)" would be fewer changes than "grow a pair of gills". $\endgroup$ – Lok Oct 25 '17 at 22:26
  • 1
    $\begingroup$ @Lok Sure that would be fewer changes, but if you don't want that then make more changes. If losing ears isn't the desired effect than you can make more changes. With GE you can make as many changes as you want, though the bigger the list the harder it will be. So yes it's less reasonable but still reasonable. $\endgroup$ – anon Oct 25 '17 at 22:52

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.