The oldest unambiguous fossil flowers date to the Early Cretaceous period, roughly 130 to 125 million years ago, when small, simple blooms began appearing alongside the ferns and conifers that had dominated land for hundreds of millions of years. But that tidy answer conceals one of the liveliest arguments in paleobotany: molecular evidence keeps pushing the origin of flowering plants deeper into time, some controversial fossils claim to be flowers from 50 million years before those Early Cretaceous specimens, and even pollen grains with suspiciously flower-like features turn up in rocks from the Triassic, over 240 million years old. The question of when the first flowers appeared turns out to be less a settled fact and more a moving target.
What the Fossil Record Shows
The most widely accepted fossil evidence places the earliest flowering plants, or angiosperms, in the Early Cretaceous. Pollen records from well-dated rock sequences in places like Portugal and northern Gondwana show simple, single-grooved angiosperm pollen grains appearing in the Valanginian stage, around 135 to 130 million years ago, with a slow increase in diversity and abundance through the Barremian and into the Albian stages that followed.1PubMed. How deep is the conflict between molecular and fossil evidence on the age of angiosperms? Portuguese coastal deposits document at least 66 different types of early angiosperm pollen across that interval, with the more advanced pollen forms restricted to younger Albian sediments.2Review of Palaeobotany and Palynology. New records of Early Cretaceous angiosperm pollen from Portuguese coastal deposits: Implications for the timing of the early angiosperm radiation
Actual fossil flowers, not just pollen, tell a consistent story. Preserved blooms from the Early Cretaceous show a steady increase in complexity over time, starting with small, structurally simple forms and diversifying into more elaborate arrangements. Many of these earliest flowering plants were herbaceous or shrubby rather than tree-sized, and monocots, the group that today includes grasses and lilies, were already diverse and prominent in some Early Cretaceous ecosystems.3PubMed Central. Diversity in obscurity: fossil flowers and the early history of angiosperms The pattern is one of slow buildup: angiosperms were around for at least 20 to 30 million years before they achieved the ecological dominance seen in some mid-Cretaceous plant communities.
When Molecular Clocks Push the Date Back
If the fossil record were the only evidence, the story would be relatively straightforward. But DNA-based dating methods tell a different tale. By comparing the genetic sequences of living plant species and calibrating the rate of change against known fossil markers, researchers have repeatedly estimated that flowering plants originated much earlier than their oldest confirmed fossils. A 2025 analysis using Bayesian dating approaches placed the origin of the angiosperm crown group squarely in the Triassic period, between roughly 255 and 202 million years ago, the same era when mammals, dinosaurs, and lizards first appeared.4PubMed Central. New insights on angiosperm crown age based on Bayesian node dating and skyline fossilized birth-death approaches
That gap of 70 million years or more between molecular estimates and the oldest well-accepted fossils is enormous by any standard. It implies that flowering plants existed for a very long time before leaving recognizable traces. One explanation is that the earliest angiosperms were rare, small, and lived in upland environments where fossilization is unlikely. Another is that molecular clocks, despite their sophistication, can be thrown off by rate variation across lineages and by the difficulty of choosing the right fossil calibration points. Earlier molecular clock studies were sometimes wildly inconsistent with the fossil record, and while newer methods have narrowed the discrepancy, a meaningful gap persists.5Evolution. THE AGE OF THE ANGIOSPERMS: A MOLECULAR TIMESCALE WITHOUT A CLOCK
The Jurassic Flower That Stirred a Fight
In 2018, a team led by Chinese researchers described a fossil called Nanjinganthus from rocks in eastern China dated to the Early Jurassic, more than 174 million years ago. Based on over 200 specimens, they argued it met the key criterion for being an angiosperm: its ovules were enclosed at the time of pollination, the defining feature that separates flowering plants from all other seed plants.6ScienceDirect / China Geology. Nanjinganthus is an angiosperm, isn’t it? If correct, this would push the origin of flowers back by at least 50 million years beyond the earliest well-accepted fossils and into the range that molecular clocks have long predicted.
The reaction from the broader paleobotanical community has been sharply divided. Critics argue that what looks like an enclosed ovule in a compressed fossil can be ambiguous, and that the flower-like features of Nanjinganthus might belong to an unrelated group of seed plants that convergently evolved similar structures. Supporters counter that the sheer number of specimens and their consistent anatomy make a strong case. The debate remains unresolved, and Nanjinganthus occupies an uneasy middle ground: celebrated by some as the oldest known flower, dismissed by others as a misidentified gymnosperm.
Even Older Hints From the Triassic
If a Jurassic flower is contentious, Triassic pollen is even more so. Pollen grains with features remarkably similar to those of early angiosperms have been found in Middle Triassic sediments, roughly 245 to 240 million years old, from northern Switzerland and the Norwegian Arctic. In Switzerland, small grains with the columnar wall structure and reticulate surface sculpture typical of angiosperm pollen were recovered from rocks of the Anisian stage.7PubMed Central. Angiosperm-like pollen and Afropollis from the Middle Triassic (Anisian) of the Germanic Basin (Northern Switzerland) Similar finds from the Barents Sea showed what researchers described as an “astonishing high diversity” of angiosperm-like pollen in Middle Triassic marine sediments.8Journal of Micropalaeontology. A boreal early cradle of Angiosperms? Angiosperm-like pollen from the Middle Triassic of the Barents Sea (Norway)
The critical word here is “angiosperm-like.” The researchers who described these grains noted that while the pollen shows all the essential structural features of angiosperm pollen, the lack of a continuous record through the rest of the Mesozoic makes it hard to connect them directly to later flowering plants. The Swiss team suggested the pollen likely came from a stem-group relative of angiosperms rather than a true flowering plant. Angiosperm-like pollen has also turned up in Upper Jurassic deposits from France, including grains closely resembling the pollen of living lilies.9Review of Palaeobotany and Palynology. Angiosperm-like pollen from the ammonite-dated Oxfordian (Upper Jurassic) of France These scattered finds are tantalizing, but they fall short of proof that true flowers existed that early. They may instead represent extinct plant lineages that evolved some flower-like traits independently.
What Makes a Flower a Flower
Part of why the debate is so heated is that drawing the line between “flowering plant” and “not a flowering plant” in the deep past is genuinely difficult. The hallmark innovation of angiosperms is the carpel, a closed structure that envelops the ovules and, after fertilization, develops into the fruit. Before the carpel evolved, seed plants produced their seeds on open, leaf-like structures exposed to the environment.10PubMed Central. Case not closed: the mystery of the origin of the carpel How and when that enclosure happened is one of the great unsolved problems in plant biology.
Extinct groups like the Bennettitales had reproductive structures that superficially resemble flowers and once led researchers to group them with angiosperms. Closer examination, however, shows that their seeds were not enclosed by a carpel or any analogous structure, and their anatomy differs substantially from both flowering plants and other proposed relatives.11PubMed. Is the anthophyte hypothesis alive and well? New evidence from the reproductive structures of Bennettitales The same problem haunts those Triassic and Jurassic pollen grains: you can have pollen that looks like it came from a flower without having a flower at all, because the wall structure of pollen can evolve independently of what the parent plant’s reproductive organs look like.
Another feature once thought to be unique to flowers is double fertilization, the process that creates both an embryo and the nutrient-rich tissue called endosperm. It turns out that a rudimentary form of double fertilization also occurs in some living gymnosperms, the Gnetales, which produce two fertilization products instead of one. Research on species like Gnetum gnemon has shown that this process was likely inherited from a common ancestor of the Gnetales and angiosperms, meaning double fertilization predates the origin of flowers themselves.12American Journal of Botany. Double fertilization in Gnetum gnemon (Gnetaceae): its bearing on the evolution of sexual reproduction within the Gnetales and the anthophyte clade In flowers, the second fertilization event was gradually repurposed from producing a spare embryo into producing the endosperm that nourishes the seed.13Sexual Plant Reproduction. The evolution of double fertilization and endosperm: an “historical” perspective
How Flowers Conquered the Planet
Whenever the first flowers appeared, they did not immediately take over. The Early Cretaceous was still a world dominated by conifers, ferns, and cycads. Angiosperms spent millions of years as a relatively minor part of the landscape before undergoing an explosive radiation in the mid-Cretaceous, roughly 100 to 80 million years ago. By the end of the Cretaceous, flowering plants had spectacularly displaced gymnosperms as the dominant vegetation around the globe.14PubMed Central. The angiosperm radiation revisited, an ecological explanation for Darwin’s ‘abominable mystery’ The transition was a true ecological revolution, reshaping terrestrial habitats and driving what researchers have called the Angiosperm Terrestrial Revolution, a prolonged shift to higher levels of biodiversity on land.15PubMed. The Angiosperm Terrestrial Revolution and the origins of modern biodiversity
Insects played a starring role. Direct fossil evidence from Burmese amber, about 99 million years old, preserves a beetle covered in the pollen of a flowering plant, providing the earliest unambiguous proof of specialized insect pollination in angiosperms.16PubMed Central. Pollination of Cretaceous flowers Other amber and compression fossils from the Early Cretaceous show that communities of flower-visiting and flower-eating insects were already forming at both local and regional levels.17PubMed Central. Florivory of Early Cretaceous flowers by functionally diverse insects: implications for early angiosperm pollination The co-evolution between flowers and their pollinators became a feedback loop: more diverse flowers supported more diverse insects, which in turn drove further floral specialization.
Herbivorous dinosaurs were part of the picture too. During the mid-Cretaceous, generalized browsing and trampling by dinosaurs may have actually favored angiosperms. Many flowering plants were small, fast-growing, and quick to reproduce, traits that gave them an advantage in disturbed habitats. As these weedy angiosperms spread, they fueled the diversification of low-browsing dinosaurs like ornithopods, creating another feedback loop between plants and animals.18Review of Palaeobotany and Palynology. The reciprocal interaction of angiosperm evolution and tetrapod herbivory
The Leaf That Changed Everything
One of the less glamorous but more consequential innovations of flowering plants was a change in their leaves. Angiosperms evolved vein densities far higher than anything seen in conifers, ferns, or other plant groups. This denser plumbing system allowed angiosperm leaves to move water and sugars more efficiently, enabling rates of photosynthesis and water loss through transpiration that no other group of land plants could match.19PubMed Central. Angiosperm leaf vein evolution was physiologically and environmentally transformative Reconstructions suggest this leaf-level innovation allowed the maximum photosynthetic capacity of angiosperms to surpass competing groups between 140 and 100 million years ago.20PubMed. Leaf hydraulic evolution led a surge in leaf photosynthetic capacity during early angiosperm diversification
The timing is significant because atmospheric carbon dioxide was declining during the Cretaceous. Under high CO₂, plants do not need to open their leaf pores as wide to absorb carbon, so the advantage of high vein density is modest. As CO₂ dropped, angiosperms with their dense leaf veins could keep their gas exchange rates high by opening their pores more freely without wilting, something gymnosperms with their lower vein densities struggled to do.21PubMed Central. A critical transition in leaf evolution facilitated the Cretaceous angiosperm revolution Modeling work supports this idea, showing that angiosperms evolved from a gymnosperm-like state of low gas exchange under high CO₂ toward much higher rates as atmospheric carbon dioxide fell.22Biogeosciences. The Cretaceous physiological adaptation of angiosperms to a declining pCO2: a modeling approach emulating paleo-traits In other words, falling CO₂ did not just happen while flowers were rising; it may have been one of the environmental pressures that gave them their competitive edge.
The Genome Duplication Question
For years, a popular hypothesis held that a whole-genome duplication, an event in which a plant’s entire set of chromosomes gets copied, occurred near the base of the flowering plant lineage and may have provided the raw genetic material for floral innovation. Whole-genome duplication has happened repeatedly across the plant family tree and has been linked to the evolution of new structures and survival through environmental upheavals.23PubMed Central. Pervasive genome duplications across the plant tree of life and their links to major evolutionary innovations and transitions
Recent work, however, has cast doubt on whether a genome duplication specifically at the base of angiosperms actually happened. A 2025 study using gene-retention patterns in early-diverging flowering plants like Amborella found that the signature of duplicated genes fits a single duplication event that predates seed plants as a whole, with no clear signal of a separate duplication at the origin of angiosperms themselves.24PubMed Central. Revisiting ancient whole-genome duplications in the seed and flowering plants through the lens of dosage-sensitive genes If this holds up, the genetic catalyst for the evolution of flowers was not a sudden doubling of the genome but something subtler, perhaps regulatory changes in existing genes that allowed old structures to be repurposed into the novel architecture of a flower.
Darwin’s Abominable Mystery
Charles Darwin famously described the rapid rise and early diversification of flowering plants as an “abominable mystery.” The phrase has been widely quoted ever since, but its original meaning is often misunderstood. Darwin was not baffled by the existence of flowers. What troubled him was the apparent speed of their appearance in the fossil record, which seemed to violate his deeply held belief that evolution always proceeds gradually. Flowering plants, among all the organisms he considered, presented the most extreme challenge to the idea that nature does not make leaps.25PubMed. The meaning of Darwin’s ‘abominable mystery’
In a way, the modern debate echoes Darwin’s frustration. If the molecular clocks are right and angiosperms originated in the Triassic, then the apparent “sudden” appearance in the Early Cretaceous is an artifact of a poor fossil record during a long cryptic phase. That would actually resolve Darwin’s complaint, because the diversification was not as rapid as it looked. If the fossil record is closer to the truth and flowers really did originate in the Cretaceous, then their subsequent explosion across the planet in a relatively short geological window remains genuinely remarkable and hard to explain by gradual mechanisms alone. Either way, the mystery Darwin identified over 140 years ago is far from solved.
When Flowers Hit the Water
One surprising thread in the story of early flower evolution is how quickly some lineages left dry land. Reconstructions of ancestral habitats suggest that the common ancestor of all living angiosperms was a land plant, but transitions from terrestrial to aquatic environments started early, nearly coincident with the origin of angiosperms in some phylogenetic models. Most of these shifts concentrated in the last 25 million years, but the earliest ones occurred far earlier in angiosperm history.26PubMed Central. Macroevolutionary dynamics in the transition of angiosperms to aquatic environments Early angiosperms may have even started as a few aquatic species before diversifying into terrestrial herbs and eventually into the towering trees, creeping vines, and desert shrubs of today.27Acta Geologica Sinica – English Edition. Mid‐Cretaceous Hothouse Climate and the Expansion of Early Angiosperms The fact that living water lilies belong to one of the oldest surviving angiosperm lineages fits this picture neatly. It hints that some of the environments most closely associated with early flowers were not forest floors but the margins of lakes and rivers, places where small, fast-growing plants could gain a foothold without competing directly against the established conifers overhead.
Chemical Ghosts of Ancient Flowers
Beyond body fossils and pollen, there is one more kind of evidence for ancient angiosperms: molecular fossils. Oleanane is a chemical compound produced today almost exclusively by flowering plants. When researchers tracked the concentration of oleanane in ancient source rocks and petroleum, they found that its abundance parallels the diversification of angiosperms from the Cretaceous onward.28PubMed. The molecular fossil record of oleanane and its relation to angiosperms Oleanane has been used as a biomarker to estimate the maximum age of source rocks, essentially as a chemical proxy for when flowering plants were common enough to leave a geochemical signature. The approach has limits: it only tells you when angiosperms were abundant, not when the first one appeared. And because it relies on the compound surviving intact through geological time, it is better at confirming presence than ruling out absence. Still, the oleanane record supports the broader picture of a Cretaceous rise, with flowering plants going from geochemically invisible to a dominant signal in the rock record over the span of about 80 million years.