The Mesozoic Era, stretching from about 252 to 66 million years ago, was dominated by plant groups that look very different from the flowering trees and grasses covering most of Earth’s land surface today. Conifers, seed ferns, ginkgos, cycads, and fern-like plants formed the bulk of terrestrial vegetation for most of that span, with flowering plants arriving relatively late and only gradually displacing the older lineages. The story of Mesozoic plant life is one of extinction recovery, gymnosperm supremacy, and a slow-motion revolution as angiosperms crept in during the Cretaceous.
Recovery After the Great Dying
The Mesozoic opened in the aftermath of the worst mass extinction in Earth’s history. The Permian-Triassic crisis, roughly 250 million years ago, wiped out an enormous proportion of life on land and in the sea. Forests were hit hard. In what is now Europe, massive dieback of coniferous vegetation degraded terrestrial ecosystems, and a prolonged period of lycopsid dominance followed for four to five million years before conifers could re-establish themselves at the transition between the Early and Middle Triassic.1PubMed Central. The delayed resurgence of equatorial forests after the Permian-Triassic ecologic crisis Lycopsids are clubmoss relatives, small-statured plants that had been major forest-builders in the Carboniferous but by this time were acting more as weedy pioneers, filling the vacuum left by the conifer collapse. The equatorial forests that had existed before the extinction took millions of years to return to anything like their previous complexity.
This sluggish recovery set the stage for the rest of the Triassic. As conditions stabilized, conifers staged a comeback. Seed ferns spread widely, especially in the southern supercontinent Gondwana. And several plant groups that would go on to define the Mesozoic landscape began diversifying rapidly.
Seed Ferns and Their Mysterious Decline
Despite the name, seed ferns were not ferns at all. They were seed-bearing plants with fern-like foliage, and they were among the most important vegetation groups for much of the Mesozoic. Several distinct orders occupied different ecological roles and different parts of the globe. One of the most iconic was Dicroidium, which was widespread across Gondwana during the Triassic but essentially absent from the northern supercontinent Laurasia.2Review of Palaeobotany and Palynology. The first record of the Gondwanan seed fern Dicroidium Gothan in Laurasia This geographic split tells us that even in a world without polar ice caps, plant distributions were shaped by continental positions, ocean barriers, and climate belts.
Another significant seed fern group, the Caytoniales, had a different trajectory. The genus Sagenopteris, a representative of this order, originated in what is now Europe during the Middle Triassic, expanded its range during the Late Triassic, and survived the end-Triassic mass extinction. It became one of the dominant gymnosperms during the Early and Middle Jurassic, then declined through the Late Jurassic and Cretaceous before its last known occurrences in mid-to-high northern latitudes during the mid-Cretaceous.3Papers in Palaeontology. Revision of Sagenopteris (Caytoniales): a major lineage of the Mesozoic seed plants That slow fade-out, coinciding with the rise of flowering plants, is a pattern repeated across many Mesozoic gymnosperm lineages.
Five orders of seed ferns have at various points been proposed as possible ancestors of flowering plants: the Glossopteridales, Peltaspermales, Corystospermales, Caytoniales, and Petriellales. The idea is appealing because some seed ferns had structures that loosely resemble angiosperm features. But the evidence remains frustratingly inconclusive, and the relationship between any of these groups and true flowering plants is still unresolved.4PubMed. Seed ferns from the late Paleozoic and Mesozoic: Any angiosperm ancestors lurking there? Seed ferns were generally thought to have vanished at the end-Cretaceous extinction alongside the dinosaurs. But fossils from Tasmania show that at least one lineage, a corystosperm seed fern called Komlopteris, survived into the Cenozoic by at least 13 million years, making it a genuine “Lazarus taxon” that emerged from apparent extinction in the fossil record.5PubMed. Seed ferns survived the end-Cretaceous mass extinction in Tasmania High-latitude regions appear to have served as refuges for relict species during global crises.
The Conifer Empire
Conifers were arguably the most ecologically important plants through most of the Mesozoic. They formed forests from the tropics to the poles, and their diversity was far greater than what we see today. Among the most dominant and widespread was the family Cheirolepidiaceae, which produced the distinctive pollen type Classopollis. These conifers were remarkably versatile, thriving in coastal environments, dry regions with low rainfall and fast drainage, hot arid climates with salt and gypsum formations, and even humid swampy areas.6PLoS ONE. Classopollis works as a significant indicator for the Cheirolepidiaceae paleovegetation arid zone, as proven by fossil records from Egypt Their Classopollis pollen turns up so abundantly in Mesozoic sediments worldwide that it serves as a reliable marker for arid and semi-arid conditions near the ancient Tethys Ocean.
What made the Cheirolepidiaceae so successful? One factor may have been polyploidy, a condition in which organisms carry extra sets of chromosomes. Analysis of aberrant Classopollis pollen from the Triassic-Jurassic boundary suggests the presence of unreduced pollen, which is one of the main routes to polyploidy in modern plants. Polyploid speciation may explain the remarkable range of growth forms and environmental tolerances seen in this conifer family, and it might have buffered them against extinction during the end-Triassic crisis.7PubMed Central. Aberrant Classopollis pollen reveals evidence for unreduced (2n) pollen in the conifer family Cheirolepidiaceae during the Triassic-Jurassic transition The Cheirolepidiaceae eventually went extinct, probably during the Late Cretaceous, as global climates shifted and angiosperms expanded. But for over a hundred million years they were among the most abundant trees on Earth.
Other conifer families that flourished during the Mesozoic include the Araucariaceae (relatives of modern monkey puzzle trees) and the Taxodiaceae (relatives of modern redwoods and bald cypress). These resin-producing conifers were responsible for much of the amber found in the fossil record. Analysis of amber chemistry from Early Cretaceous deposits in Brazil, for instance, shows resin signatures consistent with araucariacean or cupressacean conifers rather than the pine family, reflecting the different mix of conifers that dominated Mesozoic forests compared to today’s northern-hemisphere pine-dominated woodlands.8PubMed Central. Revealing the diversity of amber source plants from the Early Cretaceous Crato Formation, Brazil
Ginkgos at Their Peak
Today only a single species of ginkgo survives: Ginkgo biloba, the familiar street tree with its fan-shaped leaves. But during the Mesozoic, ginkgo relatives were spectacularly diverse. The group experienced a steep increase in diversity from the Middle Triassic onward, reaching its zenith in the Late Triassic when almost all known ginkgoalean families occurred simultaneously. In China alone, the number of genera of ginkgo-like leaf forms climbed rapidly, with origination rates above 88% for some of the most common types.9Geological Journal. The rise of ginkgoalean plants in the early Mesozoic: a data analysis Major innovations during this period included the appearance of dwarf shoots and the differentiation of leaves into a blade and a stalk, features still visible in the modern species.
Ginkgoalean diversity remained high through the Jurassic. Fossils from the Early to Middle Jurassic of the Qaidam Basin in northwestern China, for example, document ten species across three common genera (Ginkgoites, Baiera, and Sphenobaiera) from a single succession of fossil-bearing layers.10Papers in Palaeontology. Early to Middle Jurassic Ginkgoales from the Qaidam Basin, northwest China The similarity between fossil ginkgo leaves and the modern species is striking, making them a textbook case of long-term morphological stability. Chemical analysis of fossil cuticles supports this: the biochemical signature of Czekanowskia, a member of the related order Leptostrobales, is strongly similar to that of modern Ginkgo leaves, consistent with a close evolutionary relationship.11Nature Ecology & Evolution. Molecular signatures of fossil leaves provide unexpected new evidence for extinct plant relationships
Ferns, Horsetails, and the Understory
Beneath and between the Mesozoic forests, ferns and horsetails formed dense ground cover. These were not bit players. In lowland and floodplain environments, monospecific thickets of horsetails like Equisetites and large ferns like Clathropteris and Dictyophyllum spread across wide areas, forming an extensive and accessible food source for herbivorous dinosaurs, especially sauropods.12Paleobotanical research journal. The systematics and ecological role of the orders Equisetales, Filicales, and Marattiales in the Mesozoic fora of Abyek and Jirande, western Alborz Drier habitats in river deltas hosted different fern species that played a smaller role in sauropod diets. The nutritional quality of these plants actually varied with atmospheric COâ‚‚ concentration: experiments growing modern relatives of Mesozoic plants under elevated COâ‚‚ show that digestibility and metabolizable energy content change depending on both the species and the COâ‚‚ level, with horsetails generally providing the highest energy yields.13PubMed Central. Diets of giants: the nutritional value of sauropod diet during the Mesozoic
The ecological importance of ferns shows up at both ends of the Mesozoic. During recovery phases after mass extinctions, ferns were often the first colonizers. And at the very end of the era, the devastating asteroid impact that closed the Cretaceous is marked in the fossil record by a “fern spike,” an abrupt shift from diverse plant pollen assemblages to sediments dominated by just a few fern spore types. This pattern, well documented in North America and also recorded in the Southern Hemisphere, signals global deforestation followed by fern colonization of denuded landscapes.14PubMed. Indication of global deforestation at the Cretaceous-Tertiary boundary by New Zealand fern spike
When Flowering Plants Arrived
Flowering plants, or angiosperms, are so overwhelmingly dominant today that it is easy to assume they have always been the main act. In reality, they were latecomers to Mesozoic ecosystems. The oldest angiosperm-like fossils come from the Late Jurassic of China, where small fruiting axes with carpels enclosing ovules have been described as Archaefructus.15PubMed. In search of the first flower: A Jurassic angiosperm, Archaefructus, from northeast China Molecular clock analyses reject the idea that all major angiosperm lineages originated entirely within the Cretaceous, instead placing the divergence of major groups across a window from about 150 to 100 million years ago.16PubMed Central. Constraining uncertainty in the timescale of angiosperm evolution and the veracity of a Cretaceous Terrestrial Revolution
Even after angiosperms appeared, their takeover was gradual. They underwent rapid diversification and geographic expansion from the early to mid-Cretaceous, but the actual transformation of ecosystems into angiosperm-dominated communities unfolded progressively and was not complete until the Late Cretaceous.17PubMed Central. The stepwise rise of angiosperm-dominated terrestrial ecosystems For tens of millions of years, flowering plants coexisted with conifers, ginkgos, cycads, and seed ferns, gradually increasing in abundance and diversity. The shift was not a sudden replacement but a slow renegotiation of ecological space.
Insects Were Already Pollinating Before Flowers Existed
One of the most surprising findings from recent paleontology is that insect pollination did not begin with flowering plants. Gymnosperm-insect pollination partnerships were already well established during the Mesozoic, involving at least four groups of insects: thrips, lacewings, flies, and beetles. Each group represents a different pollination style linked to its particular mouthpart anatomy and feeding habits, and their host plants included ginkgos, cycads, conifers, and the now-extinct bennettitaleans.18PubMed. False Blister Beetles and the Expansion of Gymnosperm-Insect Pollination Modes before Angiosperm Dominance
Some of this evidence is spectacularly direct. Early Cretaceous amber from Spain preserved thrips covered in gymnosperm pollen grains. These thrips had specialized ring-shaped bristles interpreted as structures for collecting pollen, functionally analogous to the pollen-collecting hairs on bees, providing direct evidence of specialized gymnosperm pollination by at least 110 to 105 million years ago.19PubMed Central. Thrips pollination of Mesozoic gymnosperms Even more recently, beetles preserved in amber have been found densely covered with pollen from the extinct gymnosperm order Erdtmanithecales, revealing a pollination mode for a plant group that has no living descendants at all.20PubMed. Beetle pollination of Erdtmanithecales, an extinct lineage of Mesozoic gymnosperms
When flowering plants did begin diversifying, they initially relied on these same generalist insect pollinators. Analysis of mid-Cretaceous angiosperm pollen from the Dakota Formation shows that about three-quarters of species were insect-pollinated, with adaptations for generalist insects being the most common mode, followed by more specialized pollen-collecting insects.21PubMed Central. Early steps of angiosperm pollinator coevolution The tight mutualistic relationships between specific flower lineages and specific insect pollinators that characterize modern ecosystems likely did not solidify until after the Cretaceous-Paleogene extinction reshaped both plant and insect communities.22Trends in Plant Science. Early diversifications of angiosperms and their insect pollinators: were they unlinked?
Forests at the Poles
Perhaps the most alien-looking feature of Mesozoic plant geography, from a modern perspective, was the presence of forests well inside the polar circles. Coniferous forests extended into high latitudes where winter brought months of continuous darkness, a light regime with no modern equivalent for large trees. Fossil plant assemblages from Ellesmere Island in the Canadian Arctic, at a paleolatitude of about 75°N, record mire environments dominated by amber-producing taxodiaceous conifers alongside ginkgos, cycads, ferns, lycopsids, and angiosperms, with leaf shape analysis suggesting most of the vegetation was deciduous.23Palaeogeography, Palaeoclimatology, Palaeoecology. Palaeoecology of Late Cretaceous polar vegetation preserved in the Hansen Point Volcanics, NW Ellesmere Island, Canada
How did these polar forests survive? Atmospheric COâ‚‚ was at least double modern levels, and modeling work using experiments on living conifer relatives suggests that the elevated COâ‚‚ did not much boost photosynthesis during the continuous daylight of polar summers, but it did cut leaf and root respiration rates by roughly 50% and 25% respectively. Those lower respiratory costs more than compensated for the lack of a temperature-COâ‚‚ interaction during summer, increasing overall forest productivity.24PubMed Central. Physiological ecology of Mesozoic polar forests in a high CO2 environment In other words, the carbon savings from breathing less allowed polar conifers to grow where temperature alone might not have been the limiting factor.
Plants as Climate Recorders
Mesozoic plants are not just interesting as organisms. They are also some of the best tools we have for reconstructing ancient atmospheres. The stomata on fossil ginkgo leaves, for instance, respond to COâ‚‚ levels: fewer stomata develop when COâ‚‚ is high, because the plant needs fewer gas-exchange pores to get enough carbon. Comparisons of stomatal density in Mesozoic ginkgo species with modern Ginkgo biloba show that the fossil leaves had stomatal index values 40 to 60% lower than modern samples. When expressed as ratios, these values track the COâ‚‚ variations predicted by geochemical models of the long-term carbon cycle, confirming ginkgos as useful proxies for ancient atmospheric composition.25PubMed. Assessing the potential for the stomatal characters of extant and fossil Ginkgo leaves to signal atmospheric CO2 change
Other plant-based evidence constrains atmospheric oxygen. Some geochemical models had predicted prolonged intervals of very low Oâ‚‚ during the Mesozoic, down to 10 to 12%. But the persistent record of wildfires throughout the era, preserved as fossil charcoal in plant-bearing sediments, is incompatible with such low oxygen levels because fire cannot sustain itself below a threshold concentration.26PubMed. Limits for combustion in low O2 redefine paleoatmospheric predictions for the Mesozoic Mesozoic wildfire charcoal, combined with experiments on ignition thresholds, demonstrates that oxygen stayed high enough throughout the era to support combustion and, by extension, fire-adapted plant communities.
Bennettitales and Other Extinct Enigmas
Several major Mesozoic plant groups have no living descendants at all, making them difficult to place in any family tree. The Bennettitales, sometimes called “cycadeoids,” superficially resembled cycads but were unrelated. Biochemical analysis of their fossilized cuticles confirms this: the chemical signature of bennettitalean leaf surfaces is distinct from that of true cycads and instead clusters closely with the equally extinct Nilssoniales, supporting the hypothesis that these two groups were allies that merely looked like cycads by coincidence.11Nature Ecology & Evolution. Molecular signatures of fossil leaves provide unexpected new evidence for extinct plant relationships The Bennettitales were widespread through the Jurassic and much of the Cretaceous, and some species had flower-like reproductive structures that once led botanists to wonder whether they might be angiosperm ancestors. That hypothesis has largely fallen out of favor, but the group’s elaborate reproductive biology remains intriguing, especially given amber evidence that beetles and thrips were visiting their extinct gymnosperm relatives and carrying their pollen.
The Erdtmanithecales, known primarily from their distinctive Eucommiidites pollen, are another vanished order that left little trace besides pollen grains and the occasional amber-preserved pollen cone. Direct evidence of beetle pollination for this group, found in amber specimens with dense pollen coverings on nitiduloid beetles, reveals an insect-mediated pollination mode that contributed to structuring Mesozoic biodiversity but has no surviving analog.20PubMed. Beetle pollination of Erdtmanithecales, an extinct lineage of Mesozoic gymnosperms These extinct groups remind us that the Mesozoic plant world was not simply a less diverse version of today’s. It contained entire lineages, ecological strategies, and pollination networks that vanished and were replaced, rather than simply upgraded, by the flowering plants that inherited the Earth.