Birds, as we currently know them, build nests and lay eggs on the ground. People and things with sharp teeth live on the ground. There are also things with claws and beaks up in the air, but fewer.

I was wondering if a bird-like animal could exist that would spend its entire life in the air, and only fall to the ground upon death. I already know that birds can sleep while airborne, but I need help with:

  • Figuring out a way to explain how my birds (or bats, or whatever) evolved to never touch the ground - I lack a compelling and plausible story.
  • The food situation. Unless I also got flying plants (unlikely, given the energy equations), it'll have to be a carnivore preying upon other flying creatures, or grab animals from the ground.
  • More specifically, I'm worried about the first stages of life. Perhaps a kangaroo like-pouch for the newborn?
  • Finally, I'd prefer to make this flying beast as large as possible for story-related purposes. How big can it get and still fly around the clock, do you think?

enter image description here

(Image above: Northrop still makes the best Quetzalcoatli)

LATE EDIT: Thanks everyone for a great number of good suggestions! All your answers are appreciated, and I wish I could accept two more in this case. Besides @VilleNiemi's answer, which I accepted, I think @Ivy_Lynx's answer deserves more upvotes and I got some interesting ideas from @CortAmmon as well.

  • 2
    $\begingroup$ You could try a pufferpolyp (see also here). Disclaimer: These are my own answers. $\endgroup$ – HDE 226868 Mar 7 '15 at 14:35
  • 3
    $\begingroup$ They could land and nest on water, and never touch the ground ;) $\endgroup$ – Mikey Mar 7 '15 at 22:18
  • 1
    $\begingroup$ Maybe they live on a gas giant and there is no ground to touch? $\endgroup$ – Doug McClean Mar 8 '15 at 0:55
  • 4
    $\begingroup$ Trees are included in "ground"? $\endgroup$ – Paŭlo Ebermann Mar 8 '15 at 12:54
  • 1
    $\begingroup$ Would the back of another flying animal count as ground? $\endgroup$ – Tom J Nowell Mar 8 '15 at 22:53

10 Answers 10



Since the defining feature is not nesting on ground, the starting point should be the nesting behaviour. There should be a strong pressure to reduce time spent on the ground such as predators or competition for suitable locations with other species. This would favour a single very large egg that takes a long time to develop inside the female, but hatches fast once outside and learns to fly as soon as possible.

Once the young hatches almost instantly once on the outside, the egg shell is replaced by live birth. The young would gain the ability to grab on the back of the adult so they can be moved away from danger if necessary. Once this is instinct gained at birth, the birds can give birth directly on the backs of their mates and skip nesting entirely.

After this the birth size would start dropping as the time spent "nesting" on the back of an adult would be reasonably safe. But the size difference between learning to fly and being an adult would remain large, probably grow.


The birds would eat fish. It is a reasonably reliable food source that supports large predators. And the winds over seas are much more stable and predictable than over terrain, which would be a major factor for large birds that never land.


Adult birds would have to be quite large for this to make any sense. The actual possible size is difficult to estimate. Especially the mass. If we assume the species is communal with young adults using powered flight while full adults simply glide and allow young ones to rest on their backs in exchange for the food... Big.

This would actually make sense for a species with such odd nesting behaviour. You might end with four age categories. Nestlings that live on the backs of their elders until the grow enough to fly. The young that can fly, but are still immature sexually and learning how to catch fish. These would still rest on the backs of their elders. They'd also go from needing to be fed, to providing food for their siblings and elders. Adults would be sexually mature and responsible for reproduction and most food gathering. The elders would provide wide backs for use and help other fliers remember that trying to eat the grandchildren of something that can swallow you whole is a bad idea. They'd probably go from being able to feed themselves to needing to have food provided for them. They'd also be the ones that would know the weather patterns and fishing grounds.

The birds would be pretty social with a structure like that with complex social interactions and vocal communication.

  • $\begingroup$ Complex interactions can occur with extremely limited communication (or intelligence), so I would hardly say that's a requirement. It's quite typical of larger insect communities for example. $\endgroup$ – David Mulder Mar 8 '15 at 2:40
  • 1
    $\begingroup$ @DavidMulder True. Not sure what "requirement" refers to. But I am basically extrapolating from existing social birds. I think a set up like this would put positive selective pressure on social intelligence and ability to communicate. That is more or less what happened with humans, right? $\endgroup$ – Ville Niemi Mar 8 '15 at 10:11
  • $\begingroup$ The weight of carrying the young around could well be a problem, still a nice approach. $\endgroup$ – Tim B Mar 8 '15 at 13:17
  • 3
    $\begingroup$ A really interesting answer! Although, birds wouldn't likely eat "vinegared rice" :P - it's a common misconception that Sushi is "raw fish". It's not any one thing really, and more arguably "rice" than "fish". (You could use "Sashimi") $\endgroup$ – James Mar 9 '15 at 2:42
  • $\begingroup$ What about "really" nesting, as in, with eggs, on the back of the elders? They might have any kind of adaptation for their backs to be suitable for eggs. $\endgroup$ – jesusiniesta Mar 24 '15 at 11:36

Why not take a look at the real-life version of this: Alpline Swift

It seems that the bird lives on insects, and drinks rain-drops. They can spend up to 6 months in the air and can even sleep on the wing. I think a few modifications to this bird will give you what you're looking for

  • 1
    $\begingroup$ Wow, I never knew these even existed!! $\endgroup$ – Serban Tanasa Mar 10 '15 at 0:23
  • 1
    $\begingroup$ There's also certain species of dragonfly that spend their entire adult life in the air. They've been found in shipping lanes in the middle of the Pacific. I think they do need to land to lay eggs, and they are of course land-based during their larval stage, but from then on, they can live, feed, and even mate on the wing. $\endgroup$ – Darrel Hoffman Mar 22 '16 at 15:12

Most creature designs, if anthropomorphized, can be thought of as a balance between seizing opportunities and fearfully avoiding mishaps. As a result, most designs are very flexible, adapting to new situations as they come. What you are looking for is a fearful one, which doesn't even try to land when there's an opportunity, so I would expect the design to show the classic sign of fear: rigidity. It would have more machine-like characteristics than organic ones, as it optimizes for an environment where not touching ground is more important than anything else. This means your bird will not necessarily look like any bird you or I have ever seen. Why have hips if you never plan on using legs?

The big thing, as you mentioned, will be food. Staying close to the ground is a liability for a creature that is uninterested in landing, because it cuts down on degrees of freedom and puts the creature in reach of predators that are not afraid to leverage the ground to create a more adaptable shape. This creature is going to have to stay higher, perhaps in the jet streams or just below them (they are really gnarly winds for a living creature to deal with). I could see them functioning more like baleen whales, sifting food from the wind (like small bugs and other organic matter that got lifted high into the air). This is, of course, a very low caloric diet, so that will shape the rest of their lives, but by hanging out near places with strong updrafts like dust devils or even hurricanes occur, they may be able to eek out calories.

In the name of conserving energy, the creature would have to evolve to some pretty extreme structures. The creature would need to be almost entirely flight-related or food-related hardware. In order to support the mass of a small brain, it might grow to giant proportions in segments, much like our modern long distance fliers. As a bonus to you, this structure also suggests they should be huge, exactly as you want.

Raising young is an interesting topic. The obvious answer is to take a non-rearing approach and just have them kick off thousands of little fliers, but that seems to be in conflict with the idea of them being large creatures hanging out in regions where its difficult for smaller creatures to fly. I think child-rearing may become a necessity.

So how do we teach them to fly? Falling out of a nest is one thing. Falling a few thousand feet to the forbidden ground is another. The parent is going to have to help teach them, but how? My recommendation would be for them to have the aerial flyer's equivalent of a pouch: a set of muscular organs which can be used to create a mellow region of airflow for the young to learn in. If a youngling gets too aggressive and goes too far into the jetstream, one could use those muscles to change the shape of the airflow and drag the youngling back into safe harbor.

For a twist, it might be interesting to have this "pouch" move over time. At first, the mother will want to know as much about airflow as possible, so she would want to keep it near the back, giving her time to sense the oncoming airflow as much as possible. She has to protect them against gusts. As the young grow, she could move it forward to give them a better sense of what the jetstream feels like without compromising her own aerodynamic integrity. The chaotic flow over a wing is important to organic flying... giving up control of it in order to give more free-flow style to the young would probably be a mistake. Eventually, the young may catch a surprise pocket, and be forced to leave their mother forever. Or, the young may stick with their mother until they are right out in front of the mother, in the free jetstream, only reliant on the mother to catch them when something goes wrong. At some point, the mother may simply bank, and leave them to their lives.

  • $\begingroup$ +1, but I would recommend the newly flying 'birds' soar behind the mother/father instead of in front. $\endgroup$ – Mikey Mar 7 '15 at 22:24
  • 1
    $\begingroup$ @Mikey My theory was that a low calorie flyer doesn't get to change direction much. If something goes wrong, there would be no way to go back for the younglings. I do see advantages to soaring behind (drafting), but in front felt safer, so that's why I chose that way. $\endgroup$ – Cort Ammon Mar 8 '15 at 3:34
  • 1
    $\begingroup$ Re "Why have hips if you never plan on using legs?" My pet bird uses the same limbs as both hands and feet. If a bird with opposable grasping digits and rotating wrists were releived of the need to use them as feet, they would further evolve toward hands rather than a compromise. I think this would allow the hips to become like our shoulders: range of motion more imortant than load bearing in a particular direction. $\endgroup$ – JDługosz Apr 3 '15 at 8:36
  • $\begingroup$ @JDługosz I like that direction. It's consistent with my argument that they would lose their hips, but is far more efficient by turning those hips into something even more useful! If there's one thing we can truly say about Evolution, it's that it doesn't let anything to to waste. It finds a way to make use of everything! $\endgroup$ – Cort Ammon Apr 3 '15 at 15:29
  • $\begingroup$ Check out wikipedia, @CortAmmon. My pet is zygodactyl, I still do a double take to realize that he eats with his feet. The attachment point at the middle of the body can reach the head like our arms bending at the elbows. The swiveling wrists have interesting results as anckles: walking along a rope parallel to his facing direction, he will rotate the thumbs inward or outward to grab it. An enormous bird would not support his own weight on lightweight stick legs. $\endgroup$ – JDługosz Apr 3 '15 at 16:50

It seems that the view on the food problem usually looks at the sea, probably because we think of albatross and the like. To me, one of the most beautiful leaps of imagination on this question is Stephen Baxter's Evolution, in which he thinks of an air whale, a creature feeding on air plankton (a true thing actually, there are swarms of insects really high up).

A summary is given on this page:

For example, Baxter invents an air whale -- a huge but very light creature that flies above most of the clouds, in the stratosphere. The air whale evolved from pterosaurs, getting lighter and lighter. Its bones are hollow, and it doesn't need a big brain since nothing much happens up there. It feeds on aerial plankton, and mates on the highest mountain peaks when its instincts tell it to. With wings one hundred meters across, it would have been magnificent to see.

However, it does land to nest.


I think any answer to this question really hinges on how earth-like you want your planet to be and how earth-like you want the rest of your fauna and flora to be. Since you don't place any restrictions on these, I'll take the liberty to assume these factors are tweakable in order to get our perpetually airborne creatures.

The other answers so far are on the right track in my opinion, so this will be more supplementary than a different idea.


The problem of explaining the evolution of a flier that never lands is twofold (I'm not a biologist :P):

  1. Why not land?
  2. Why remain in the sky?

It seems like the same question but it isn't.

Why not land?

On the first point, you can avoid most land without being in the air (there's oceans, tall mountains, caves etc.). The usual dangers aren't sufficient: predators can be avoided through changing location, changing your place on the food cycle or changing the hours you look for food. Thus, there is something far more dangerous or detrimental that makes landing a guaranteed darwin award. Those that land die or don't have babies (or their babies don't survive). This could be because of starvation, predation etc. but I've already excluded these as insufficient dangers. It has to be something ubiquitous on land which is harder to adapt to than flying forever.

I can't really think of anything ubiquitous enough that wouldn't be dangerous to all other life as well. It might be some temporary situation that lasted long enough for the adaptation of permanent flight (or just not ever landing on ground) but subsided far enough into the past to allow other life to re-emerge, leaving the fliers still in the air. Perhaps when both factors are combined there is sufficient motive to both avoid land and keep to the sky. Taking this into account, it may be possible to make predators a likely motivator - perhaps in the distant past, birds and fliers where so ubiquitous that land and water predators made them their main source of food, eventually making the ground and sea so dangerous that some of these creatures evolved to avoid that danger entirely.

Why remain in the sky?

In order to remain in the air perpetually, you would most definitely require an ecosystem. Whether that includes or excludes plants, it would definitely require at least one or two other species that have also adapted to perpetual flight. Have the mythical finally arrived?

Indeed, it seems Gilles' idea and the ancient worldbuilding tag may be the answer. If you have large balloon-whales evolve to slowly and perpetually roam the skies, they could be the symbiotic platform to allow landing without grounding. By virtue of being full of hot air, they could remain aloft while allowing comparatively smaller creatures to nest on their backs (or somewhere safer from the wind, unless the whale has some sort of triceratops head providing cover). These need not evolve from actual whales, just be creatures that can survive on air plankton and sunlight. They could have the temperament of sloths rather than whales, consuming air-plankton, insect and fumes as they float around (or swim in the air), taking advantage of strong currents to help them.


The creatures that nest on them would need to provide them with something, so perhaps they are small herbivores or omnivores (since meat is more efficient food) that spend the majority of their time in the air, but feed from the oceans or forest tops. Their droppings and leftovers are used by the balloon-whales to supplement their diet and energy requirements (or protein requirements, whatever).

Given this arrangement, your large never-landing creatures could be sky-sharks. Like sharks, their offspring are born ready to fly and hunt. They feed on the smaller air-dwelling creatures, particularly by snatching them as they descend for food, in droves. With a sufficiently large wing-span, carnivorous diet etc. they could possibly sustain themselves and rest while flying. Many sea creatures sleep while swimming and some birds can sleep while flying so this isn't that far-fetched.


At this point, you can add sky-wolves etc. to create a more complete ecosystem (as well as other birds that regularly land as usual).


At this point the answer is already quite clear: balloon-whales are born floating, the goblin-hawks on their backs are born on the backs of the whales and can grow strong enough for flight, assisted or not, before they make their first attempts and the sky-sharks are, like sharks, born ready.

Of course, you could have offspring be born in flight and only ready enough to climb on to the parent's back and cling their until they're strong enough (which would require a long and flexible neck on the parent's side in order to be able to feed it, which is convenient here). As they cling, they learn.

Pouches would work as well.


Unfortunately I'm way too sleep deprived to get into research and math here, so I'll just make some wild guesses.

A large wingspan would most likely be required, hence the sky-sharks would have to be large. How large exactly would depend on atmospheric properties, the amount of food they get and how high they fly (I think). However, if we allow for some creative adaptations, perhaps their size can be increased.

In particular, we know that planes are massive and heavy, but can still fly just fine and even glide. The requirement is speed (as far as I understand). If they can find enough strong winds, drafts and currents, have enough of a wingspan to take advantage of these to lose as little speed as possible and have some form of propulsion to compensate, it may be achievable.

Propulsion could take the form of air sacs that fill up as they fly and are used as an afterburner. Or they could expel bowel gases. Considering we already have assumed balloon-whales that fly with the power of hot air, it isn't nearly as ridiculous as it would otherwise be to assume that the sky-sharks have a similar adaptation. Some sea-creatures use water propulsion (as in, not just swimming through water) and I think some microbial life does as well, so it is possible at least in theory fantasy. Perhaps they ignite it too :P.



So here's where we can start: Some birds (megapodes) are superprecocial, i.e. they can fly on the day they hatch. So you need a bird that can do this, but also give birth to live young. They could then be born in mid air and glide while the wings dry out. Other parts of the animal kingdom have species which lay eggs fairly closely related to those with live young (e.g. snakes, fish), so we could assume that live birth could evolve in your "birds" (massive oversimplification, but I guess we're not really discussing descendents of current earthly birds).

After that if you want small, they could eat insects -- aerial birds such as swifts stay aloft except when breeding and live on insects. Larger insects there were dragonfly-like insects with a 65cm wingspan in the carboniferous could give larger birds.

Bigger birds would still have to take food on the wing. They could perhaps live on the small, swift-like birds in my previous paragraph, or on conventional birds. Large birds over land tend to use thermals or updraughts to get lift (e.g. vultures); for this to work you could do with cliffs or terrain giving a strong temperature contrast -- a shoreline might work. Over the sea it's possible to gain lift from local wind differences -- I've read something more recent but can't find it online.

Size: a prehistoric bird Argentavis magnificens had an 8m wingspan! Unfortunately no-one was around to record its behaviour. The current largest-winged bird (the wandering albatross) feeds at night and can therefore be assumed to be able to fly at night. If they use wind-shear- or wave-slope-soaring this will keep them aloft at night, as both techniques depend on the wind and not the sun. Soaring would appear to be a requirement for sustained flight of a heavy bird (also Condors).

Your birds wouldn't be able to weigh much though -- the heaviest living flying bird is (or might be) the great bustard at only about 20kg, while A. magnificens was estimated to weigh up to 72kg -- the same as a fairly small adult human. Birds' hollow bones are obviously a big help in keeping the weight down.

How it evolved: live birth is tricky but it's evolved multiple times on earth -- just not in birds. Think about nest predators -- fast, agile but non-flying creatures that eat eggs and flightless young. Monkeys maybe. These would have to be quite fearless to snatch the eggs from under an incubating parent -- or perhaps devious (dismantle a nest from below), or armed with something dangerous enough to threaten the adult's ability to incubate or look after young. If you combine chronic pressure for nest sites, a quicker and quicker gestation may be selected for.

A simpler solution would of course be to have the bird nest somewhere high up out of the reach of predators.

  • 1
    $\begingroup$ It could also have an egg pouch, where the mother carries eggs around with her until they hatch. These will also act as nests for the hatchlings, until they get big enough to fly on their own $\endgroup$ – nzaman Jul 12 '16 at 17:40

There is actually a videogame where you need to escort an egg that has exactly such a bird in it, Amerzone.

The bird in that game is even born in the air. The egg is basically a pouch with a thermal sail that hatches above a live volcano. The birds fall out, develop wings as they fall, and fly away, eating, mating and dying in the thermal drafts above the volcanoes of the Amerzone area. Their wings are said to keep growing after they die, and they are said to keep flying even after death.


In Saturn Rukh, there is no ground so life forms had to evolve that way. The Rukh was large, as you can see in the cover depicted next to the human's space ship.

Rukh cover

It has two brains, one for day and one for night. It is a filter feeder, and follows its food sources.


I created a species of flying creatures a few years ago. They're like a mixture of blimps and manta rays. They are shaped like manta rays, and have an air sack they fill with gases like helium using special gill like filters. They use the appendages by their mouths to catch other flying prey.

These creatures also fly in swarms, with the females in the middle. The babies grow on the mother's belly in a special clear sack. This means the females have larger air sacks; so they can carry their babies. The babies learn how the males protect the females, see their parents catch food, etc. Once they are fully developed and ready to fly, they either puncture the sack, or spit acid on it to dissolve it. I haven't decided which. Life goes on (for everyone but the pufferpolyps).


Yes there is, The common swift (Apus apus) is a medium-sized bird, superficially similar to the barn swallow or house martin but somewhat larger. It is, however, completely unrelated to those passerine species, since swifts are in the separate order Apodiformes. The resemblances between the groups are due to convergent evolution reflecting similar life styles. The swifts' nearest relatives are thought to be the New World hummingbirds and the Southeast Asian treeswifts.

Their scientific name comes from the Ancient Greek words α "without", and πούς, "feet". ἄπους, apous, meaning "without feet". These birds have very short legs which they use only for clinging to vertical surfaces (hence the German name Mauersegler, literally meaning "wall-glider"). They never settle voluntarily on the ground, where they would be vulnerable to accidents and predation.

  • 2
    $\begingroup$ While this is interesting, a vertical surface would count as "ground" for the purposes of this question. $\endgroup$ – Brythan May 10 '16 at 17:59

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.