Evolutionary convergence is one of the most fascinating phenomena in evolution. It shows that species from very distant lineages, separated by hundreds of millions of years, can develop strikingly similar traits. These similarities, presented as the result of separate evolutions, raise the question of whether the invoked mechanisms are really sufficient to explain such close solutions.
Swimming fast to escape a predator, perceiving light in the depths, conquering the air, resisting temperature variations, optimizing gas exchange in low-oxygen environments, withstanding extreme pressures, saving energy during scarcity, detecting chemical signals, or camouflaging to avoid predation: all these are constraints imposed by the environment.
According to theory, the evolutionary mechanism, lacking plan, purpose, or design, explores and selects, generation after generation, the most effective solutions. But when very distant lineages seem to arrive at the same answers, again and again, the question arises: does nature really reinvent, or does it follow more marked paths than we think? It’s as if evolution is pushed, almost against its will, to repeat some of its inventions.
This phenomenon seems to challenge the idea of evolution as entirely contingent and random. It would no longer be just chance, but a form of recurrent tendency: natural selection, like an inflexible engineer, pushes species toward a limited set of effective solutions. As if the laws of matter and energy impose, in the background, certain possible forms of life. Wings for flying, eyes for seeing, venoms for neutralizing prey… so many innovations that have appeared in unrelated lineages, to the point of suggesting that nature might follow an instruction manual that we only glimpse.
The paleontologist Stephen Jay Gould (1941–2002), who saw evolution as a largely unpredictable and contingent process, nevertheless acknowledged that convergence raises a troubling question: what if certain forms of life are not just the product of chance, but also the expression of deeper, perhaps inevitable constraints? In his essay Wonderful Life, he imagined “replaying the tape of life” and wondered: if evolutionary history started over, would the same creatures, or at least the same solutions to the challenges of existence, reappear? Convergence seems to suggest repetition… but also leaves doubt about what in evolution is truly random.
This idea opens a dizzying perspective: the diversity of life, despite its apparent profusion, might fit into a more predictable framework than we imagine. The constraints of the physical world would then act as a filter, allowing only certain innovations to emerge—those that, curiously, end up appearing in species with very distant origins. A lesson in humility, perhaps, but also an invitation to wonder if there are not deeper, still poorly understood rules that guide nature’s inventiveness.
| Supposed convergent trait | Said independent groups | Adaptive function | Age of divergence | Example & particularity |
|---|---|---|---|---|
| Single-lens eye | Vertebrates / Cephalopods | Image formation | > 550 Ma | Human (blind spot) / Cuttlefish (no blind spot)—striking similarities despite a shared genetic heritage |
| Active flight (wings) | Insects / Pterosaurs / Birds / Bats | Aerial movement | > 350 Ma | Four different body plans, but aerodynamic constraints seem to impose similar solutions |
| Carcinization (crab shape) | True crabs / False crabs (anomurans) | Protection, lateral walking | > 200 Ma | “Crabification”: distant lineages adopt the same silhouette, perhaps under similar constraints |
| Echolocation | Bats / Dolphins / Swiftlets | Navigation and hunting in darkness | > 80 Ma | Similar acoustic mechanisms, but sometimes surprisingly convergent genetic bases |
| Camouflage (homochromy) | Insects / Birds / Mammals | Concealment | > 300 Ma | Stick insect, orchid mantis, Arctic hare—similar strategies, but not always independent biochemical pathways |
| Paralyzing venom | Snakes / Slow worms / Spiders / Scorpions / Frogs | Immobilization of prey or defense | > 200 Ma | Similar toxins, sometimes based on gene families already present in the common ancestor |
| Bioluminescence | Insects / Deep-sea fish / Bacteria / Fungi / Squid | Attraction, lure, camouflage | > 500 Ma | Luciferin and luciferase: same chemical reactions, but truly independent origins remain debated |
| Extremophily (radiation resistance) | Archaea / Bacteria / Tardigrades | Extreme resistance | > 2.7 Ga | Similar DNA repair mechanisms, perhaps derived from very ancient molecular modules |
Among all the examples of evolutionary convergence, the eye is undoubtedly the most spectacular, and perhaps also the most puzzling. The more biologists explore its origins, the more the idea of simple, obvious convergence seems to complicate.
The common ancestor of octopuses and humans was a primitive bilaterian worm, living about 550 to 600 million years ago (late Precambrian). This animal did not have a structured eye with lens, retina, and pupil, but it was not entirely devoid of visual tools: it already had rudimentary photoreceptors, the Pax6 gene, the true conductor of ocular development, as well as an ancestral opsin shared by the entire animal lineage.
This is precisely where the notion of convergence becomes less obvious. When Walter Jakob Gehring (1939–2014) showed in 1983 that injecting the mouse Pax6 gene into a fruit fly could trigger the formation of an eye on an antenna, he revealed that the genetic program underlying the eye is common to animals separated by hundreds of millions of years. The octopus and the human may not have “reinvented” the eye from scratch: they seem to have reactivated, each in their own way, a common ancestral molecular toolkit, resulting in remarkably similar architectures.
Atavism refers to the reappearance, in a modern organism, of an ancient trait that disappeared millions of years ago. This phenomenon shows that some traits are not truly “reinvented” by evolution: they can simply be reactivated from ancestral genetic programs still present in the genome, but silent. Chickens developing teeth, whales reappear with legs, or snakes showing limb buds illustrate the ability of life to reactivate ancient structures rather than recreate them.
Atavism thus complicates the interpretation of evolutionary convergence: what is sometimes presented as an independent invention may simply be a reactivation of a pre-existing genetic module. In other words, some “convergences” may not be new solutions imposed by environmental constraints, but possible returns to ancient states inscribed in the deep history of life.
The framework proposed by Charles Darwin (1809–1882) in 1859 is based on three major principles: descent with modification from common ancestors, heritable variation among individuals, and natural selection as a sorting mechanism. These three pillars are now confirmed by genetics, paleontology, molecular phylogeny, and direct observation. No serious member of the scientific community questions them.
Darwin grasped the essentials, even if the mechanisms he proposed are no longer sufficient to explain everything. 21st-century biology does not contradict his work: it extends, completes, and enriches it, sometimes beyond what was imagined.
Evolutionary convergence shows how much the environment and physical laws can steer evolution toward similar solutions, without being entirely left to chance. It thus opens the possibility—but not the certainty—that forms of extraterrestrial intelligence could emerge elsewhere, if similar constraints apply. Eyes, wings, or analogous social structures could then appear… or suggest that life sometimes follows more marked paths than we imagined.
But this theoretical possibility should not obscure a deeper reality: evolutionary trajectories remain largely shaped by unique, unpredictable sequences.
And yet, despite these troubling similarities, terrestrial life remains the product of an extraordinarily contingent history, shaped by a succession of improbable and non-reproducible events. Nothing suggests that this path, with its breaks, accidents, and detours, could be reproduced elsewhere; everything suggests, on the contrary, that it is unique to Earth and its singular history.