Humans lost their tails roughly 25 million years ago, when the ancestors of all living apes split from Old World monkeys. The loss was not gradual whittling over millions of generations but appears to have been triggered by a single genetic mutation, identified in a 2024 study, that altered how a key developmental gene gets read during embryonic growth. What makes this story stranger is that every human still grows a tail in the womb, only to reabsorb it before birth. The genetics, the embryology, and the evolutionary pressures behind this change are more tangled and more interesting than a simple “we didn’t need it anymore” explanation.
Every Human Grows a Tail Before Birth
By the fourth week of gestation, a human embryo has a clearly visible tail. It extends beyond the developing legs and, at its peak, makes up roughly a tenth of the embryo’s length. This is not a metaphorical bump or a vague protrusion. It is a genuine tail, complete with vertebrae precursors, and it looks remarkably similar to the tail of a mouse embryo at a comparable developmental stage.
The tail starts disappearing around the eighth week. The primary mechanism is programmed cell death: cells in the tail tissue essentially receive a signal to self-destruct in an orderly fashion. Researchers confirmed this using methods that detect dying cells, finding that the patterns of cell death in the human embryonic tail closely mirror those seen in mouse embryos. The process depends on a type of programmed cell death called caspase-dependent apoptosis, and it appears to be the dominant way tissue is removed during tail regression in both species.1PubMed Central. Spinal neural tube formation and tail development in human embryos
A key molecular event happens just before the tail begins to shrink. Two growth-signaling molecules in the tail bud get switched off at a specific embryonic stage, followed by a burst of cell death that wipes out the progenitor cells responsible for continued growth. The timing of this shutdown differs somewhat between humans and rodents, but the basic sequence of events is shared.2eLife. Spinal neural tube formation and tail development in human embryos In other words, human embryos do not simply lack a tail program. They run a full tail-building program and then actively dismantle the result.
The Mutation That May Have Started It All
For decades, nobody could point to the specific genetic change responsible for tail loss in apes and humans. That changed in 2024, when a team led by Bo Xia at New York University identified a strong candidate. The culprit is a short DNA element called an Alu element that inserted itself into a gene called TBXT (historically known as Brachyury, or the T gene). TBXT is one of the most important genes for tail development across vertebrates; when it malfunctions in mice, their tails are shortened or absent.
The Alu insertion landed inside a non-coding stretch of TBXT in the common ancestor of all hominoids. Once there, it paired with another, older Alu element already sitting nearby but facing the opposite direction. This pairing created a loop in the genetic message that causes part of the gene to be skipped when it is read, producing an altered version of the TBXT protein. The researchers found this alternative version of the protein in apes and humans but not in monkeys or other tailed primates.3PubMed Central. On the genetic basis of tail-loss evolution in humans and apes
When the team engineered mice to carry the same altered version of TBXT, many of those mice were born with shortened or absent tails. That is about as close to a smoking gun as evolutionary genetics gets. It does not prove that this single mutation was the only factor, and the researchers were careful to note that other genetic changes may have reinforced or fine-tuned the trait over millions of years. But it strongly suggests that one insertion event in one gene was enough to kick off the process of tail loss in the hominoid lineage.
There was a cost. Some of the engineered mice also developed neural tube defects, a category of birth defects that includes conditions like spina bifida. This raises the possibility that the same mutation that freed our ancestors from their tails also slightly increased the risk of certain developmental problems, a trade-off that natural selection evidently judged worthwhile.
The Molecular Network That Controls Tail Length
TBXT does not work alone. Across mammals, the final length and composition of the tail depend on a tug-of-war between growth-promoting and growth-stopping signals in the tail bud. Research in mice has identified a regulatory network involving two main players. On the growth-promoting side, a molecular pathway encourages progenitor cells to keep dividing, extending the tail. On the growth-stopping side, a group of genes acts to shut down those progenitors and halt elongation.4PubMed Central. The vertebrate tail: a gene playground for evolution In mammals, the spinal cord tissue within the tail is also actively removed at late developmental stages through the action of the growth-stopping genes, which is why our coccyx contains no spinal cord at all.
Another piece of the puzzle involves a signaling molecule that helps coordinate where the trunk ends and the tail begins. In mouse experiments, this molecule was shown to work through a chain of downstream targets that control how the embryo’s body regions are assigned along its length.5Developmental Cell. Gdf11 Signaling Coordinates the Trunk-to-Tail Transition through Retinoic Acid and Isl1 and Controls Bodily AP Regionalization Disrupting these signals in lab animals produces bodies with the wrong proportions: trunks that are too long, tails that are too short, or vice versa. The point is that tail length is not governed by a single on/off switch. It emerges from a network of interacting genes, any one of which can be tweaked by evolution to produce a shorter or longer tail. In hominoids, the TBXT mutation appears to have been the decisive tweak, but the broader network set the stage.
Why Tails Are So Useful for Other Primates
If you want to appreciate what our ancestors gave up, look at how monkeys use their tails. For many species, a tail is not a decorative appendage but an active tool for staying alive in the trees.
Studies of small-bodied monkeys moving along branches show that they use their tails as dynamic counterweights, sweeping them in the direction of an imbalance to generate a corrective rotational force. Different species use this mechanism in different ways. Tamarins, for instance, swing their tails through large arcs at high speed, actively generating the momentum needed to stay on top of a branch. Squirrel monkeys tend to hold their tails in a more fixed, depressed posture, using them as a passive counterbalance rather than an active whip.6PubMed. Tail function during arboreal quadrupedalism in squirrel monkeys (Saimiri boliviensis) and tamarins (Saguinus oedipus)
A broader study of five primate species found that all of them used forearm rotation to shift their body weight while walking along branches, but the tail-whip balancing mechanism was used most heavily by the largest and most ground-dwelling species in the sample. The researchers interpreted this as evidence that larger primates, being less naturally agile in trees, rely more on their tails as auxiliary stabilizers.7PubMed. Maintenance of above-branch balance during primate arboreal quadrupedalism: coordinated use of forearm rotators and tail motion That finding makes the loss of the tail in apes, a group that includes several large-bodied species that still spend significant time in trees, all the more puzzling.
Perhaps the most vivid demonstration of tail utility comes from wild capuchin monkeys in Brazil. Capuchins transport heavy stones and other objects while walking on two legs, and their tails turn out to be essential for this task. Despite constituting only about five percent of their body mass, the tail provides enough counterbalance to allow bipedal carrying that would be physically impossible without it.8PubMed. The strategic role of the tail in maintaining balance while carrying a load bipedally in wild capuchins (Sapajus libidinosus): a pilot study This raises the question: if a tail helps with bipedal balance, and our ancestors were in the process of becoming bipedal, why did they not keep theirs?
What Upright Posture Changed
The most commonly offered explanation is that the shift from a horizontal trunk posture to an upright one made tails less useful and potentially even a liability. All living apes, whether they walk on two legs or not, hold their torsos relatively upright compared to monkeys. This upright trunk orientation is associated with a suite of anatomical changes: a broader, flatter rib cage, more mobile shoulders, a shorter and stiffer lower spine, and a pelvis reshaped for weight-bearing from below rather than from behind.9PubMed. Evolution of the hominoid vertebral column: The long and the short of it
In a pronograde (horizontal-bodied) animal, a tail serves as a counterweight and steering mechanism. In an orthograde (upright-bodied) animal, the center of gravity sits directly above the hips, and balance depends more on the arrangement of the spine, pelvis, and lower limbs than on a trailing appendage. Once the hominoid ancestor began spending more time with its trunk upright, whether during climbing, sitting, or suspensory locomotion, the tail’s role as a balancing organ would have been diminished. A long tail dragging behind an upright body could even be a metabolic cost with no payoff: tissue to grow, blood to supply, injuries to risk, and parasites to host.
It is worth noting, though, that this explanation has limits. No one has shown experimentally that a tail would be purely disadvantageous for an upright primate. The capuchin evidence described above actually suggests that tails can aid bipedal balance. The honest assessment is that the shift to upright posture reduced the selective pressure to keep a tail, and the TBXT mutation happened to arise in a lineage already trending in that direction. The mutation was not lethal, the tail was no longer critical, and so the trait stuck.
What the Coccyx Actually Does
People sometimes describe the coccyx (tailbone) as a “useless leftover,” but that is not quite right. The coccyx is the fused remnant of several tail vertebrae, and while it no longer supports a protruding tail, it has been repurposed. It serves as an anchor point for several muscles, ligaments, and tendons of the pelvic floor. It also forms one point of a three-legged structure, along with the two sitting bones, that supports your weight when you sit down.10PubMed Central. Coccydynia: an overview of the anatomy, etiology, and treatment of coccyx pain
This is a pattern that shows up repeatedly in evolution: a structure’s original function fades, but it gets co-opted for something new. The coccyx no longer wags, but it does structural work. Calling it “vestigial” in the strict sense, meaning functionless, is misleading. A more accurate description is that it is vestigial with respect to tail function but functional in its current anatomical role. The distinction between vestigial structures and ones that have been genuinely repurposed is often glossed over in popular accounts of evolution, and the coccyx is a textbook example of why it matters.
Babies Born With Tails
Rarely, a human baby is born with what appears to be a tail. These cases have been documented in the medical literature since at least the 1800s, and they continue to generate attention because they seem to represent evolution in reverse. The reality is more nuanced, and the medical distinction between a “true tail” and a “pseudotail” matters.
A true human tail is a soft, skin-covered protrusion from the lower back. It is composed of fat, connective tissue, muscle bundles, blood vessels, nerves, and sensory receptors, but it does not contain bone, cartilage, or spinal cord. It arises from the retention of structures that all human embryos possess but normally eliminate. True tails can be as long as 13 centimeters, can move and contract in response to stimulation, and occur about twice as often in males as in females. They are surgically removed without complications and are almost never inherited within families.11PubMed. Human tails and pseudotails
A pseudotail, by contrast, only looks like a tail from the outside. It is actually caused by an underlying abnormality such as an elongated coccyx, a fatty tumor, or another developmental anomaly. Unlike true tails, pseudotails are frequently associated with spinal malformations that require more involved medical management.12PubMed Central. A True Human Tail in a Neonate: Case report and literature review The clinical importance of this distinction is real: a true tail is a harmless curiosity, while a pseudotail can signal a problem that needs imaging and follow-up.
True tails are sometimes described as atavisms, meaning the reappearance of an ancestral trait that was lost through evolution. This distinguishes them from vestigial structures, which are features that persist in reduced form in an entire species. An atavism is a one-off throwback in an individual; a vestigial structure is a species-wide remnant. The coccyx is vestigial (everyone has one), while a baby born with a tail is atavistic (it happens only occasionally).13Anatomical Science International. Atavistic and vestigial anatomical structures in the head, neck, and spine: an overview
Why the Tail Did Not Come Back
Once a complex structure like a tail is lost, re-evolving it is extraordinarily difficult. Tails require coordinated development of vertebrae, muscles, nerves, blood vessels, and skin. The genetic instructions for building all of those components in a tail-specific arrangement have been degrading for 25 million years in the hominoid lineage. Random mutations have been accumulating in genes that were once essential for tail function but are no longer under selective pressure to stay intact. Getting all of those genes back into working order simultaneously is, for practical purposes, impossible.
This is a general principle in evolutionary biology sometimes called Dollo’s law: complex traits, once lost, do not return in their original form. There are occasional exceptions for simple traits, but nothing as anatomically involved as a vertebrate tail has ever been convincingly shown to re-evolve after being fully lost in a lineage. The rare babies born with true tails are not re-evolving a tail; they are failing to complete a demolition program that still runs in every embryo. The blueprint echoes are still there, but the construction crew is long gone.
The persistence of the embryonic tail itself is, in some ways, the most revealing clue. Embryonic development tends to be deeply conservative: it is easier for evolution to add a new shutdown step at the end of a developmental sequence than to rip out the entire sequence and rebuild the embryo’s growth plan from scratch. So we still build the scaffolding of a tail every time a human embryo develops, and then we tear it down. The TBXT mutation and the apoptosis program work together to ensure that, by the time you are born, the only trace left is a small triangular bone at the base of your spine and the muscles that attach to it.