The Permian extinction, roughly 252 million years ago, killed an estimated 80 to 90 percent of all marine species and about 70 percent of terrestrial vertebrate species, making it the worst mass die-off in Earth’s history. What made it through reads like a list of organisms that happened to hold the right biological cards for a world turned hostile: animals with efficient oxygen-transport systems, creatures that could burrow or enter dormancy, fast-reproducing lineages that could bounce back from population crashes, and plants tough enough to tolerate heat, drought, and damaged soil. The survivors were not simply lucky, though luck played a role. Their biology intersected with a specific set of lethal stressors in ways that gave them a narrow edge.
What Made the Extinction So Deadly
To understand what survived, you need to know what was doing the killing. The trigger was massive volcanism from the Siberian Traps, a flood-basalt province that released enormous quantities of carbon dioxide, sulfur dioxide, and other gases over tens of thousands of years. The downstream effects cascaded: global temperatures spiked, the oceans lost dissolved oxygen (a condition called anoxia), seawater became more acidic, and evidence from sulfur isotopes points to severe disruption of the ozone layer, which would have bathed land surfaces in damaging ultraviolet radiation.1PubMed Central. Respiratory protein-driven selectivity during the Permian-Triassic mass extinction Soil erosion from dying forests dumped nutrients into coastal waters, feeding microbial blooms that consumed still more oxygen, creating a vicious feedback loop.2PubMed Central. Centennial scale sequences of environmental deterioration preceded the end-Permian mass extinction
The combination was devastating because it attacked life from multiple angles simultaneously. Marine organisms faced hot, oxygen-poor, acidified water. Land animals faced extreme heat, thinned atmospheric oxygen, and UV stress. Plants faced acid rain, drought, and ozone damage. Any organism that was vulnerable on even one of those fronts was in danger; organisms vulnerable on several had almost no chance.
The Oxygen Bottleneck in the Oceans
In the sea, the single biggest predictor of whether a group survived or went extinct was its ability to handle low-oxygen conditions. A study using multiple logistic regression analysis across marine invertebrate genera found that the end-Permian extinction selected against organisms with poorly buffered respiratory physiology and calcareous (calcium-carbonate) shells.3Geology. Acidification, anoxia, and extinction: A multiple logistic regression analysis of extinction selectivity during the Middle and Late Permian In plainer terms, if a marine animal did not have an efficient way to grab oxygen from depleted water, and its skeleton dissolved in acidic conditions, it was in deep trouble.
Recent work has sharpened this picture by looking at the respiratory proteins different groups carry. Organisms whose blood chemistry included hemoglobin or similar oxygen-binding proteins had measurably better survival odds than those relying on simpler gas-exchange systems. The interpretation is straightforward: when dissolved oxygen plummeted, the ability to actively bind and transport what little oxygen remained became a matter of life and death. Warmer water made this worse by simultaneously holding less dissolved oxygen and raising an animal’s metabolic demand for it.1PubMed Central. Respiratory protein-driven selectivity during the Permian-Triassic mass extinction Ocean acidification, driven by the same volcanic carbon release that caused the warming, compounded the crisis for anything that built a shell out of calcium carbonate.4The Innovation. Respiratory protein-driven selectivity during the Permian-Triassic mass extinction
One wrinkle that complicates a clean story: you might expect burrowing shellfish, which routinely deal with low-oxygen sediment, to have been pre-adapted for anoxic conditions. They were not. Infaunal bivalves (those living within the sediment) were actually especially likely to go extinct, even though they should have been best adapted to tolerate oxygen-poor, carbon-dioxide-rich environments.5PubMed Central. Hyperthermal-driven mass extinctions: killing models during the Permian–Triassic mass extinction The extinction, in other words, was severe enough to overwhelm even organisms with some built-in tolerance for one stressor, because the combination of stressors was unprecedented.
Marine Winners and Losers
Brachiopods, which had dominated the seafloor for hundreds of millions of years, were hit catastrophically hard. They look superficially like clams but are anatomically very different, and their physiology left them poorly equipped for the new conditions. Bivalves, by contrast, fared much better and moved into ecological roles brachiopods once held. A global paleoecological study of shell beds documented a dramatic, geologically sudden takeover by bivalves as the dominant organisms on level-bottom marine surfaces during the earliest Triassic, a dominance that continued for millions of years.6Paleobiology. When bivalves took over the world For a long time, researchers debated whether bivalves actively outcompeted the weakened brachiopods or simply filled the vacuum. A more recent Bayesian analysis found that bivalves did not drive brachiopods out through competition; rather, brachiopods had been hit so much harder by the extinction itself that bivalves expanded into empty space.7PubMed Central. Bayesian analyses indicate bivalves did not drive the downfall of brachiopods following the Permian-Triassic mass extinction
Ammonoids, free-swimming relatives of modern nautiluses, tell a different story. They were hit badly during the extinction itself, but their characteristically high rates of reproduction and speciation allowed them to bounce back with startling speed. Within about two million years of the boundary, Triassic ammonoids had already reached diversity levels higher than in the Permian.8PubMed. Good genes and good luck: ammonoid diversity and the end-Permian mass extinction This explosive recovery contrasted sharply with bottom-dwelling groups like bivalves and gastropods, which took far longer to rebuild their diversity. The difference likely comes down to life-history strategy: ammonoids reproduced quickly and speciated fast, so they could repopulate available niches almost as soon as conditions stabilized, even if only briefly. Both ammonoids and conodonts (tiny eel-like marine animals with hard tooth-like structures) experienced multiple crises during the Early Triassic but rediversified dramatically between each one, thanks to those same high turnover rates.9PubMed Central. Evidence from ammonoids and conodonts for multiple Early Triassic mass extinctions
Free-swimming marine animals had another advantage over bottom-dwellers: mobility. Nektonic organisms like conodonts and fish could migrate toward cooler or more oxygenated waters as conditions deteriorated, while sessile organisms cemented to the seafloor had to endure whatever local conditions delivered.10Journal of Oceanography and Marine Research. Solving the Mystery of Endless Life between Conodonts and Lampreys, Plus a Reason for Final Extinction of the Conodonts
Lystrosaurus and the Terrestrial Vertebrate Story
On land, the poster child for Permian-Triassic survival is Lystrosaurus, a pig-sized, tusked herbivore belonging to the dicynodont therapsids, a group more closely related to mammals than to reptiles. After the extinction, Lystrosaurus became absurdly common: in some Early Triassic fossil assemblages, it accounts for the vast majority of vertebrate specimens. Its success was not based on a single trait but on a suite of features that happened to match the demands of a post-apocalyptic world.
First, Lystrosaurus was likely warm-blooded. Its nasal cavities contained complex scroll-like structures (maxilloturbinals) positioned directly in the respiratory airflow, a feature associated with thermoregulation and moisture retention in modern mammals and birds. It also had a barrel-shaped chest that accommodated relatively large lungs, giving it an edge in extracting oxygen from an atmosphere that may have been depleted.11Darwin’s Door. Darwin’s Door Second, bone histology studies show that dicynodonts including Lystrosaurus grew extraordinarily fast compared to other therapsids, reaching adult size quickly. Enlarged vascular channels in their bone appear to have allowed especially efficient nutrient assimilation. Reaching reproductive maturity sooner would have been a major advantage when population crashes were frequent and environments unstable.12Zoological Journal of the Linnean Society. Do extraordinarily high growth rates in Permo-Triassic dicynodonts (Therapsida, Anomodontia) explain their success before and after the end-Permian extinction?
Third, Lystrosaurus was a burrower, which gave it access to underground food sources and shelter from surface temperature extremes. And there is evidence from tusk growth rings in Antarctic populations that it could enter torpor, a state of reduced metabolic activity similar to hibernation. This flexible physiology, the ability to ramp metabolism up or down depending on resource availability, may have been the clincher. In a world where food supply, temperature, and atmospheric oxygen were all swinging wildly, an animal that could essentially “power down” during bad spells and resume activity when conditions improved had a tremendous survival advantage.13Communications Biology. Evidence of torpor in the tusks of Lystrosaurus from the Early Triassic of Antarctica
The Rise of Archosaurs
The mammal-line therapsids like Lystrosaurus dominated the immediate aftermath, but the long-term beneficiaries of the extinction were the archosauromorphs, the lineage that includes crocodilians, pterosaurs, and dinosaurs. Following the extinction, archosauromorphs diversified into a remarkable variety of body plans, habitats, and locomotor styles. They filled ecological roles on land, in freshwater, and eventually in the air, showing a burst of evolutionary experimentation during the Triassic that laid the groundwork for the dinosaur-dominated Mesozoic.14PubMed Central. Locomotion and the early Mesozoic success of Archosauromorpha A study of cranial shape disparity showed that non-archosaurian archosauromorphs achieved surprisingly high morphological diversity during the Middle Triassic, which then declined as archosaurs proper (the ancestors of dinosaurs and crocodilians) expanded their own range of body forms through the Late Triassic.15PubMed Central. Unappreciated diversification of stem archosaurs during the Middle Triassic predated the dominance of dinosaurs
Freshwater as a Refuge
Freshwater environments appear to have served as critical havens during and after the extinction. Temnospondyls, a diverse group of amphibian-like animals, survived the extinction in part because large bodies of water tend to stay cooler than surrounding land, buffering aquatic residents against the extreme surface temperatures. Their aquatic lifestyles, combined with generalist diets, allowed temnospondyls to persist at low latitudes where fully terrestrial vertebrates could not. Freshwater ecosystems also offered a more stable food base compared to devastated terrestrial landscapes, where resources were sparse and unpredictable.16PubMed Central. The ecology and geography of temnospondyl recovery after the Permian–Triassic mass extinction
In the oceans, a similar geographic logic applied. Silica-secreting organisms like radiolarians and siliceous sponges saw their diversity gradients dramatically rearranged. After the extinction, silica productivity retreated to high-latitude and deep-water settings that functioned as thermal refugia, cooler pockets where temperatures and oxygen levels remained more tolerable.17Paleoceanography and Paleoclimatology. Response of Siliceous Marine Organisms to the Permian‐Triassic Climate Crisis Based on New Findings From Central Spitsbergen, Svalbard For organisms unable to migrate, being in the right place geographically was as important as any physiological advantage.
The Collapse and Reshuffling of Plant Life
The Permian extinction ended one of the most enduring plant biomes in Earth’s history. In the southern supercontinent of Gondwana, the dominant glossopterid gymnosperms, which had formed vast forests for tens of millions of years, went extinct. The immediate aftermath was characterized by a collapse in plant productivity. Opportunistic fungi, algae, and ferns flooded into the vacuum.18GSA Bulletin. Refined Permian–Triassic floristic timeline reveals early collapse and delayed recovery of south polar terrestrial ecosystems
The famous “fungal spike” recorded in sediments around the world marks this moment of ecological collapse. When the forests died, dead wood and organic debris accumulated on a massive scale, and fungi proliferated to decompose it. Palynological records show the disappearance of gymnosperm-dominated pollen assemblages, replaced first by fungal remains and carbonized plant debris, then by marine microfossils, and only later by lycopod spores and eventually new gymnosperm pollen as vegetation slowly recovered.19Geology. Fungal event and palynological record of ecological crisis and recovery across the Permian-Triassic boundary
The plants that did survive and eventually came to dominate were those adapted to hotter, drier conditions. Fossil pollen records from the Tethys region document a marked increase in conifer abundance and diversity, along with a sharp decline in glossopterids and cordaites. Peltasperms and drought-tolerant lycopsids also expanded, indicating a landscape shifting toward warmer, semi-arid conditions.20Geological Journal. Tracing the End‐Permian Floral Transition Through Palynofossils From the Gungri Formation, Spiti Basin, Tethyan Himalaya In short, the plants that made it through were the xerophytes, the drought-tolerant, heat-resistant lineages that had been marginal players in the lush Permian but were ideally suited for the degraded Triassic world.
Insects and Small Survivors
Insects are sometimes overlooked in discussions of the Permian extinction because their fossil record is patchier than that of marine invertebrates or large vertebrates. But recalibrated evolutionary timelines using well-preserved Middle Triassic fossils suggest that several major insect lineages, including the ancestors of butterflies, true bugs, and flies, originated earlier than previously thought and persisted through the extinction. Key evolutionary innovations like powered flight and complete metamorphosis appear to have evolved well before the Permian-Triassic boundary, which may have given these lineages enough ecological flexibility to weather the crisis.21PubMed Central. Recalibration of the insect evolutionary time scale using Monte San Giorgio fossils suggests survival of key lineages through the End-Permian Extinction
Among the smallest organisms, cyanobacteria and green algae were early colonizers of devastated environments. High-resolution analyses of post-extinction sediments reveal successive waves of algal proliferation, starting with simple single-celled forms and progressing to more complex charophyte algae. Coinciding with these algal blooms, concentrations of amorphous organic matter produced by photosynthetic bacteria surged, jumping from pre-extinction levels of around 2 to 11 percent to 33 to 41 percent in the early post-extinction phase.22Geology. Lethal microbial blooms delayed freshwater ecosystem recovery following the end-Permian extinction Cyanobacterial blooms also occurred in coastal seas following the vegetation collapse on land, as nutrient-laden soil washed into the water, creating conditions that favored simple photosynthesizers over more complex eukaryotic algae.23Global and Planetary Change. End-Permian terrestrial disturbance followed by the complete plant devastation, and the vegetation proto-recovery in the earliest-Triassic recorded in coastal sea sediments These microbial communities were not just survivors; they were the first rebuilders of primary productivity in a world stripped of complex ecosystems.
The Lilliput Effect
Surviving species did not simply carry on at their pre-extinction size. Across virtually all animal groups, body sizes shrank markedly after the extinction and stayed small through the entire Early Triassic, a phenomenon known as the Lilliput effect. Quantitative analyses confirm that this was not just a sampling artifact: both marine invertebrates and terrestrial vertebrates show genuine size reductions that lasted for millions of years before body size began to creep back up.24Palaeogeography, Palaeoclimatology, Palaeoecology. The Lilliput effect in the aftermath of the end-Permian extinction event Among therapsids on land, statistical tests show significant decreases in body size in the earliest Triassic compared to the Late Permian for both therocephalians and cynodonts, two groups that produced the eventual ancestors of mammals.25PLOS ONE. Body Size Reductions in Nonmammalian Eutheriodont Therapsids (Synapsida) during the End-Permian Mass Extinction
Why would getting smaller help? Smaller animals need less food and less oxygen per individual. In a resource-depleted world where atmospheric oxygen may have been lower and food webs were shattered, being small meant lower metabolic overhead. Smaller organisms also tend to mature faster and reproduce sooner, which matters enormously when populations are crashing and rebuilding repeatedly. The Lilliput effect is really the biological signature of a world selecting hard for r-strategy traits: fast growth, early reproduction, small body size, and generalist diets.
Why Recovery Took So Long
One of the more striking features of the Permian extinction is how long it took ecosystems to recover. For most marine groups, full recovery to pre-extinction diversity levels required roughly five to ten million years, far longer than for any other mass extinction. Researchers working on Early Triassic sections, particularly in South China, have identified three interlocking reasons. First, extreme environmental conditions did not simply snap back to normal after the initial volcanic pulse; intermittent volcanism and continued greenhouse warming kept the planet hostile for millions of years. Second, the survivors were predominantly fast-reproducing, ecologically simple organisms (the r-strategists) rather than the complex, specialized, slow-growing species (K-strategists) that form the backbone of mature ecosystems. Third, the resulting ecosystems were immature and fragile, lacking the structural complexity to withstand further environmental shocks.26Global and Planetary Change. Events during Early Triassic recovery from the end-Permian extinction
Ammonoids are a revealing exception. Their recovery was explosive rather than gradual, reaching pre-extinction diversity within about two million years, while most bottom-dwelling marine groups lagged far behind.8PubMed. Good genes and good luck: ammonoid diversity and the end-Permian mass extinction The difference underscores how much recovery speed depended on a group’s intrinsic biology. Fast reproducers with wide geographic ranges could rebound quickly between environmental crises; slow-growing specialists with narrow niches could not.
Each Extinction Is Its Own Kind of Filter
It is tempting to extract general rules about what survives mass extinctions, and researchers have tried. But machine-learning analyses trained on one extinction event generally fail to predict which groups survive a different extinction. When algorithms were trained on end-Permian selectivity and tested on the end-Cretaceous, or vice versa, prediction accuracy dropped to near-chance levels for most combinations.27PubMed Central. Oversimplification risks too much: a response to ‘How predictable are mass extinction events?’ The Permian extinction favored organisms that could handle low oxygen and acid water. The end-Cretaceous extinction, caused by an asteroid impact, favored organisms that could survive darkness and cold. The traits that saved you in one scenario might be irrelevant or even harmful in another.
There is also a sampling problem that shapes how we understand survival and extinction in the fossil record. Rare species, the ones that show up in only a small fraction of sampled intervals, tend to disappear from the fossil record well before the actual extinction boundary, creating the illusion that the extinction was gradual when it may have been abrupt. Statistical tests for this bias at the Permian-Triassic boundary support an abrupt extinction scenario, with the apparent pre-boundary decline being an artifact of incomplete preservation rather than a real ecological signal.28Geology. Evidence for abrupt latest Permian mass extinction of foraminifera: Results of tests for the Signor-Lipps effect This matters for understanding survivors: some lineages that appear to have vanished before the main extinction pulse may have actually persisted right up to the boundary, meaning the true survivor pool could be both larger and more sharply defined than the raw fossil record suggests.
The Ozone Connection and Land Plants
One underappreciated kill mechanism, at least for terrestrial life, was the destruction of the ozone layer. Sulfur isotope data from the key Meishan boundary section in China show anomalous values that cannot be explained by ordinary chemical processes and instead point to intense UV-driven photolysis of volcanic sulfur dioxide in the upper atmosphere. Producing and preserving those isotopic signatures required prolonged, severe disruption of atmospheric ozone, meaning the Earth’s surface was bombarded with elevated UV radiation right at the onset of the extinction.17Paleoceanography and Paleoclimatology. Response of Siliceous Marine Organisms to the Permian‐Triassic Climate Crisis Based on New Findings From Central Spitsbergen, Svalbard This helps explain the widespread destruction and mutation of land plants seen in the record, and it adds another dimension to why burrowing animals and underground plant structures fared better: they were physically shielded from the UV assault.
For organisms living in exposed terrestrial environments, ozone loss would have compounded the stresses of heat, drought, and poor air quality. Underground roots, rhizomes, and seeds would have been partly protected, which aligns with the pattern of spore-producing plants like ferns and lycopsids being among the first colonizers of post-extinction landscapes. These plants reproduce via tough-walled spores that can tolerate UV exposure better than delicate seeds or exposed reproductive structures. The combination of UV tolerance, small stature, fast reproduction, and an ability to grow in degraded soils made them the botanical equivalent of Lystrosaurus: the right set of traits for a ruined world.