Bats evolved to fly through a gradual transition from tree-dwelling ancestors that likely glided between branches before developing powered, flapping flight. This transformation required simultaneous changes in skeletal structure, wing membranes, genetics, and metabolism, and it happened roughly 55 to 52 million years ago during the early Eocene epoch. No other mammal has pulled off this feat, making bats the only mammals capable of true powered flight.
What Bats Evolved From
Bats belong to a large group of mammals called Laurasiatheria, which also includes carnivores, horses, whales, and hedgehogs. Genetic evidence consistently places bats as one of the earliest branches within this group, diverging before carnivores, hoofed animals, and their relatives split apart. Despite decades of molecular study, pinpointing the single closest non-flying relative of bats has proven difficult. Different analyses have suggested links to carnivores, to hoofed mammals, or to hedgehogs and shrews, but none of these relationships is firmly settled.
What researchers do agree on is that the ancestor of all bats was a small, nocturnal, tree-dwelling mammal. Living in trees is considered an optimal starting point for evolving controlled flight, because an animal that already climbs and leaps between branches is primed to benefit from any membrane or body shape that slows a fall or extends a jump.
The Oldest Bat Fossils
The earliest confirmed bat fossils come from early Eocene deposits dating to roughly 56 to 48 million years ago. The oldest known complete bat skeletons were found in the Green River Formation of southwestern Wyoming, dating to about 52.5 million years ago. One of the most important of these is Onychonycteris finneyi, a small bat weighing around 40 grams (about the weight of a few stacked quarters) that displays the most primitive combination of features known for any bat.
Onychonycteris still had claws on all five fingers of its forelimbs, something no living bat retains. This suggests it was an agile climber that spent significant time gripping branches. Its wing proportions fell just outside the range of modern bats and close to the range of living gliding mammals like flying squirrels and sugar gliders. Aerodynamic modeling shows it had low flight efficiency compared to any bat alive today, consistent with an animal in the early stages of powered flight rather than one that had fully mastered it. Isolated teeth from slightly older deposits in North America hint that bat-like animals may have existed even earlier, possibly extending back into the late Paleocene.
From Gliding to Flapping
The leading model for how bat flight evolved proposes a stepwise transition from gliding to powered flapping. Researchers have modeled this by creating four intermediate body plans, starting with a form resembling a flying squirrel (short fingers, a small wingspan of about 16 centimeters, and a simple membrane between the body and limbs) and ending with a form closer to Onychonycteris (elongated fingers and a wingspan of 24 centimeters).
The key breakthrough was the elongation of the hand. When researchers modeled Onychonycteris without its elongated outer fingers, its aerodynamic profile closely matched that of living gliding mammals. Adding the elongated hand back into the model produced a 2.3-fold increase in aspect ratio (the wing’s length relative to its width) and a 28% decrease in wing loading (how much body weight each unit of wing area has to support). Those two changes together dramatically improved flight capability, turning a glider into something that could sustain flapping.
This suggests that the earliest bat ancestors were gliders with membranes between their limbs and body, much like sugar gliders today. The rapid elongation of the fingers then transformed those gliding membranes into true wings capable of generating thrust.
How the Wing Was Built
A bat’s wing is structurally very different from a bird’s. Instead of feathers attached to a fused arm, a bat flies on a thin, elastic membrane stretched across four enormously elongated fingers. The thumb stays short and retains a claw for gripping, while the four other fingers extend to remarkable lengths. In the common bent-wing bat, the longest finger is 1.54 times the length of the entire head and body combined. The thumb, by contrast, is only about 9% of head-body length.
The wing membrane itself is divided into distinct sections, each with a different role. The plagiopatagium stretches between the body’s flank and the fifth finger and serves as the structural foundation for flight. A smaller membrane called the propatagium runs from the shoulder to the wrist, while the dactylopatagium fills the spaces between the elongated fingers at the wingtip. Many bats also have a tail membrane called the uropatagium connecting the legs and tail. Embedded within these membranes are specialized muscles, including ones in the plagiopatagium and propatagium, that give bats fine control over wing shape during flight. During embryonic development, the plagiopatagium forms through a process where skin from the trunk grows outward and fuses with both the forelimbs and hindlimbs, a process that begins midway through fetal development.
The Genetics Behind Longer Fingers
The dramatic finger elongation that makes bat flight possible comes down to changes in gene activity during embryonic development. A growth-signaling protein called Bmp2 plays a central role. In developing bat embryos, Bmp2 is produced at levels roughly 31 to 35% higher in the hand bones of the forelimb compared to the same bones in mice. This increase stimulates cartilage cells to multiply faster and mature more fully, directly driving bone elongation.
Experiments on bat embryo limbs confirmed this: adding Bmp2 protein to developing forelimb digits made them grow even longer, while blocking Bmp2 with an inhibitor stunted their growth. Importantly, closely related signaling proteins (Bmp4 and Bmp7) showed no significant increase in bat hands compared to mice, and several other growth-plate genes looked identical between the two species. This points to a very targeted evolutionary change: not a wholesale rewrite of limb development, but a regulatory tweak that cranked up the output of one specific gene in one specific location.
A family of genes called Hox genes also contributes. In bat embryos, five of these genes remain highly active in the elongating hand bones of the forelimb long after the equivalent bones in the feet have stopped responding to them. This sustained Hox activity helps keep the finger bones growing well past the point where they would normally stop in other mammals.
Rewiring Metabolism for Flight
Growing long fingers and wing membranes is only half the challenge. Flapping flight is one of the most energy-intensive forms of movement in the animal kingdom. A bat in flight burns energy at a rate three to five times higher than the maximum a similarly sized ground-dwelling mammal can sustain during exercise. Crossing that metabolic barrier required evolutionary changes at the cellular level, particularly in the mitochondria, the structures inside cells that generate energy.
Mitochondria produce about 95% of the energy currency (ATP) that muscles need for movement. A study comparing bat genes to those of other mammals found strong signs of adaptive evolution in the genes encoding the mitochondrial energy-production machinery. About 23% of the energy-production genes encoded in mitochondrial DNA showed evidence of positive natural selection along the ancestral bat lineage, compared to only about 1% of unrelated nuclear genes. Nuclear genes involved in energy production also showed elevated rates of adaptive change, at about 5%.
The changes were concentrated in the molecular complex responsible for the final step of energy production, the one that actually assembles ATP. Three of the five positively selected nuclear genes, plus one mitochondrial gene, all belonged to this same complex. This pattern suggests that natural selection specifically targeted the efficiency and output of energy production in the ancestors of bats, enabling them to sustain the extraordinary metabolic demands of flapping flight.
Did Echolocation or Flight Come First?
One of the longest-running debates in bat evolution is whether echolocation evolved before, after, or alongside flight. Onychonycteris, the most primitive known bat, complicates the picture. Its cochlea (the inner ear structure used for processing sound) was relatively small, which initially led researchers to conclude it could not echolocate. That would support a “flight first” scenario. However, later analysis of its skull bones found possible connections between structures involved in producing echolocation calls, suggesting it might have had some echolocation ability after all. The skull was badly crushed during fossilization, making a definitive answer impossible from that specimen alone.
An alternative “echolocation first” hypothesis proposes that early bat ancestors used ultrasonic calls to detect flying insects from a stationary perch, snatching them with already-elongated forelimbs. In this model, the arms and fingers were already getting longer for climbing and reach-hunting before they became wings, and gliding and then flight developed later as the animals began leaping from perches to chase prey at greater distances. Some researchers have even suggested that echolocation ability traces back to the common ancestor shared by bats and shrews, both of which use some form of ultrasound. The question remains genuinely unresolved, and answering it will likely require the discovery of older, better-preserved fossils.