I'm creating a sort of fantasy world that I'm still trying to keep scientifically accurate (it's more alternate history than anything since I'm having it take place in our world with a few points of divergence) and I want to justify the evolution of mermaids from hominids, which gives me a time frame of only about 6 million years, a lot of questions regarding mermaid evolution have already been answered so I'm really just looking for some justification that it could happen so quickly.

edit: My mermaids are tool users so they still have the evolutionary incentive to maintain a relatively human upper body.

  • 1
    $\begingroup$ At least you got the timescale right... $\endgroup$
    – Trish
    Apr 21, 2022 at 9:28
  • $\begingroup$ Closely related: Realistic sea humanoids $\endgroup$ Apr 21, 2022 at 10:42
  • $\begingroup$ The fossil record suggests that evolution doesn't always occur at a slow, fixed speed. By whatever mechanism and for whatever reasons, organisms can evolve quickly when under intense pressures to do so. For the species that survive, at least. I'm not sure you need to justify it, or that "millions of years" is a particularly short time period. In Greg Bear's Darwin's Radio, humanity gives birth to an entirely new human species in a single generation. $\endgroup$
    – John O
    Apr 21, 2022 at 16:13
  • $\begingroup$ @JohnO you may naturally select the best swimmers from every generation but it's still going to be many many generations before they become mermaids. $\endgroup$
    – user253751
    Apr 21, 2022 at 18:08
  • $\begingroup$ @user253751 There is evidence that significant evolution has occurred in startling few generations. Punctuated equilibrium and all that. In Greg Bear's words, badly paraphrased, "even evolution is evolving". While it seems highly improbable that this could happen in a single generation as it did in his fiction, I don't think this takes millions of years. That's more than enough time to turn a wolf-like thing into whales. $\endgroup$
    – John O
    Apr 21, 2022 at 18:55

2 Answers 2


You're not the first person to ask this sort of question. To crib from one of my own answers to one of those other questions:

Your timescale seems pretty brief, but might be doable. Australopithicines diverged from the rest of the hominins about five to eight million years ago. Looking at the evolution of cetaceans, five million years is enough time to get from something that's basically a land-dwelling quadruped mammal that can dive for food to an obligate aquatic and unambiguously whale-like animal with all of the major physiological changes along the way that would require. Similarly, the ancestors of modern seals evolved from an otter-like Puijila to a very seal-like Pteronarctos over a similar timescale, though the changes weren't quite as dramatic as those that whales underwent.

Ending up with something that looks like a classical mermaid is another matter altogether, but there has been plenty written about them and your question doesn't go into any details about the nature of your mermaids, so I won't make any assumptions there.

Unlike KerrAvon I'm not quite so down on the mermaid bodyplan, though their objections aren't unreasonable. Many human groups have successfully managed to hunt and forage in the sea (most notably the Ama divers of Japan) so the human upper body and head is not entirely useless so long as you can use tools. Having a lower body that's largely useless on land hasn't hurt seals and walruses and sealions too much, either. That said, if your mermaids aren't tool users, then something much more seal or manatee like would be the more likely evolutionary outcome.

  • $\begingroup$ I would add one element: external pressure to evolve. Simply being able to fish better isn't enough. Perhaps a new species of giant predatory bird could emerge and gradually become an apex predator, filtering out any large species that can't seek refuge under water. As long as our mermaids can't avoid leaving the water at some point during their life, the bird doesn't go extinct and continues to exert pressure on the species to become fully aquatic. $\endgroup$
    – Hene
    Apr 22, 2022 at 12:41
  • $\begingroup$ @Hene why isn't it enough? exploiting a new environmental niche seems to have worked OK for proto-seals and cetaceans... I'm not aware of any suggestion that either was chased into the sea by eg. phorusracids. A lot of evolutionary stuff just kinda happens, as is the nature of random changes. $\endgroup$ Apr 22, 2022 at 12:50
  • $\begingroup$ Yes, but there has to be a reason for the old species to disappear. For example, evolving fins comes at the cost of land mobility, so there has to be a reason for that drawback to become irrelevant, otherwise the new evolutionary branch just dies out for being less adapted to normal land life. (The reason for the old species disappearing is often just the new species, but in this case they're in different habitats.) $\endgroup$
    – Hene
    Apr 22, 2022 at 13:00
  • $\begingroup$ @Hene history seems to have taken care of australopithecus well enough without the need for a super-predator, and the intermediate species of aquatic humanoid would presumably have gone the same way for the same reason... outcompeted (or hybridized and outbred or whatever else) by a smarter or more environmentally-suited descendant. $\endgroup$ Apr 22, 2022 at 13:06


The thing about evolution is that it is the result of organisms with favourable mutations having an increased chance of surviving long enough to reproduce, and the next generation likewise and so on. The problem with mermaids (assuming that the term refers to both male and female individuals with a human-like body and head but a fish-like lower torso and tail) is that the upper and lower portions of the body are favourable mutations for completely different environments.

Let's look at the upper body first. There is a head which naturally orients to look in a plane perpendicular to the axis of the spine, with both eyes spaced apart but looking in the same direction to give good depth perception but poor all-around vision. The arms and hands are adequate for climbing in conjunction with legs, good for tool use, and combine with the eyes to make humans the best throwers of rocks and spears on the planet.

Now let's look at the lower half of the body. It's a fish-like tail. It can presumably propel the mermaid through the water fairly quickly, and it is completely useless for anything else.

Put the two together and you have a creature that is fairly bad at everything. In the water:

  • the neck needs to bend at an awkward angle in order for the creature to see forwards when it is swimming
  • the mouth and nose are badly positioned for breathing without breaking out of the water completely (there's a good reason that whales and dolphins have blowholes and other mammals have nostrils at the front of the head, not the underside)
  • the body and head are unstreamlined, with arms that are poor at underwater swimming and cannot effectively throw objects underwater.

On land - well, it can flop about a bit, but the arms are unable to propel it effectively due to being too far forward of the centre of mass and the head is again at the wrong angle to see where it's going without straining. It's climbing-and-throwing optimised arms are pretty useless with a fluked tail down below. The only real advantage of the hands, arms and head arrangement is that the mermaids would have the tool-using ability to make seashell bikini tops, which only helps them survive predation by modern film censors.

Put another way - if the hominids need to spend more and more time in the water, eventually becoming almost fully aquatic then they are likely to eventually end up looking like dolphins or dugongs (the latter having been suggested as a possible source of mermaid myths). It is only if they stay on land that the upper body and head will look "human", but then they must have the legs to go with it.

As for timeframe - you are quite correct that 6 million years is not very long for such a radical change, it is about how long it took humans and chimpanzees to diverge from a common ancestor (5-10 million years). Changing from having two distinct legs to a fluked tail is pretty much what happened with the evolution of the Sirenians, which took close to 50 million years. In order to knock an order of magnitude off the required time there would need to be very strong environmental pressure (at just the right rate of change to be survivable by evolving) combined with an increased rate of mutation, whether as a result of radiation, chemical pollutants or viral action. Given that most mutations are disadvantageous, greatly increased reproductive rate and shortened generations would be necessary - which unfortunately means that the mermaids probably aren't the sharpest tools in the shed intellectually in addition to their other disadvantages.

In short - the timeframe is barely manageable by ramping up the environmental pressure and mutation rate, but unfortunately I cannot see any justification for the contradictory upper and lower body elements being a survivable evolutionary path.

  • $\begingroup$ The fact that the upper body isn't adapted to water isn't a major problem. As long as it is not actively detrimental, perhaps nature just hasn't gotten around to optimizing it yet. It does offer the enormous advantage of tool use, which may well outweigh all of its disadvantages, just like how our massive brains outweigh the resulting birth issues. $\endgroup$
    – Hene
    Apr 22, 2022 at 12:37
  • 1
    $\begingroup$ @Hene the issue is that there has to be such enormously strong environmental pressure to be optimally aquatic that only mutations that lead to no legs and a single tail survive, rather than lesser adaptations like broader and/or webbed feet and a tail that assists with swimming (eg beaver, crocodile) but doesn't replace the legs. Tool use is really good, but it is much easier to build tools on land because that's where there is access to fire and less resistance to every tool movement. $\endgroup$ Apr 22, 2022 at 15:02

You must log in to answer this question.

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