The first true fruit on Earth was almost certainly a small, dry pod that split open to release its seeds, produced by one of the earliest flowering plants sometime during the Mesozoic Era. Pinning down exactly which species bore it and when remains one of botany’s most contested puzzles, with fossil candidates pushing the date back as far as the Early Jurassic, roughly 174 million years ago. What we do know is that the fruit was not a peach or a berry or anything you would recognize at a grocery store. It was a modest structure formed by a folded leaf-like organ called a carpel, and its invention changed life on land forever.
What Makes a Fruit a Fruit
In botanical terms, a fruit is the mature ovary of a flowering plant, the structure that encloses and protects the developing seeds. It originates from the carpel, and sometimes other floral parts contribute to it as well.1International Journal of Science and Technology Research Archive. Fruit and seed evolution in angiosperms Before the carpel evolved, seed plants (gymnosperms like conifers, cycads, and ginkgoes) produced their seeds on open, exposed structures. Seeds just sat there on leaf-like surfaces, unprotected, at the mercy of the elements.2PubMed Central. Case not closed: the mystery of the origin of the carpel The carpel changed everything by folding closed around the ovules, sealing seeds inside a protective chamber. That closure is what defines a flowering plant, or angiosperm, and once the seeds mature inside that chamber, you have a fruit.
This distinction matters because plenty of non-flowering plants produce fleshy, fruit-looking structures. The squishy orange coating around a ginkgo seed and the red cup that surrounds a yew seed both look and function like fruits in the sense that animals eat them and disperse the seeds. But anatomically they are not fruits. They lack the defining innovation: an enclosed ovary derived from a carpel.
Fleshy Structures Before True Fruits
One of the more surprising findings in plant evolutionary biology is that the molecular toolkit for making fleshy, attractive seed coverings existed long before flowering plants appeared. Researchers studying the ginkgo and the yew found that both species use the same types of MADS-box genes to build their fleshy seed structures that angiosperms use to ripen their fruits. The ginkgo’s fleshy outer seed coat and the yew’s aril develop from completely different tissues, yet they recruit overlapping genetic programs.3PubMed. Molecular analyses of MADS-box genes trace back to Gymnosperms the invention of fleshy fruits This means that the gene networks for producing soft, seed-dispersing tissues were assembled independently in distantly related lineages of gymnosperms, well before angiosperms appeared on the scene.
The yew’s ripening process turns out to be more similar to what happens in an angiosperm fruit than the ginkgo’s is, which is interesting because the yew’s aril is a structure that forms from scratch around the seed rather than from a pre-existing seed coat. The implication is that nature stumbled on fleshy seed coverings multiple times through convergent evolution, reusing similar genetic raw materials. By the time the first true angiosperm fruit appeared, the concept of wrapping seeds in something appealing to an animal was not exactly new. What was new was the carpel-based architecture that made the whole system far more versatile.
The Fossil Candidates
Identifying “the first fruit” in the fossil record is complicated by the rarity of well-preserved plant reproductive structures and by fierce debate over what counts as an angiosperm. Two fossils have attracted the most attention.
The first is Archaefructus, discovered in Upper Jurassic deposits in northeastern China. Its fossilized fruiting axes carry small follicles, each formed from a folded (conduplicate) carpel arranged in a spiral. Elongated stigmatic crests are visible on each carpel, and the ovules are clearly enclosed, satisfying the botanical definition of a fruit.4PubMed. In search of the first flower: A jurassic angiosperm, archaefructus, from northeast china When it was first described in 1998, Archaefructus was hailed as the oldest known flowering plant, potentially over 140 million years old. Its fruits were dry and small, consistent with what botanists expected the earliest angiosperm fruit to look like.
The second, even more controversial candidate is Nanjinganthus, from Early Jurassic rocks in China, roughly 174 million years old. Researchers initially reported ovules enclosed within ovaries based on comparisons of over two hundred specimens, but critics pointed out a nagging problem: the ovules were only visible in broken ovaries, and no one had confirmed that an intact ovary actually contained ovules. More recently, micro-CT scanning resolved the issue by demonstrating both enclosed ovules and a closed ovary wall in the same specimens.5Scientific Reports. Micro-CT results exhibit ovules enclosed in the ovaries of Nanjinganthus If Nanjinganthus holds up as a genuine angiosperm, it would push the origin of fruits back by tens of millions of years. Many paleobotanists remain skeptical, but the micro-CT evidence has strengthened the case.
When Did Flowering Plants Actually Originate
The fossil record gives us individual snapshots, but molecular clock studies try to estimate when the entire group of flowering plants first diverged. Recent analyses using Bayesian dating methods and fossil-calibrated models estimate that the crown group of angiosperms originated during the Triassic period, roughly 255 to 202 million years ago, overlapping with the first appearances of dinosaurs, mammals, and lizards.6PubMed Central. New insights on angiosperm crown age based on Bayesian node dating and skyline fossilized birth-death approaches If accurate, this would mean the earliest fruits existed well before the oldest fossils we have found so far, which is not unusual. Soft plant tissues fossilize poorly, and the first angiosperms were likely small, inconspicuous plants in habitats where preservation was unlikely.
There has been a long-standing gap between what molecules suggest and what rocks show. Darwin famously called the apparently sudden appearance of flowering plants in the Cretaceous fossil record an “abominable mystery.” The Triassic estimates help close that gap by suggesting angiosperms were present much earlier than the Cretaceous burst that fills the fossil record, just rare and hard to find. The first fruits, then, may have appeared in a world dominated by conifers, ferns, and cycads, a world that looked nothing like the flower-rich landscapes that would follow.
What the Earliest Fruits Probably Looked Like
Forget anything plump or sweet. The earliest angiosperm fruits were almost certainly tiny and dry. Fossil evidence from Early Cretaceous deposits consistently shows that both seeds and fruits were small throughout most of that period.7PubMed. Seed Size, Fruit Size, and Dispersal Systems in Angiosperms from the Early Cretaceous to the Late Tertiary The plants producing them were herbs or small shrubs growing in the understory or in semi-open woodland, not towering trees with heavy, dangling fruits.
The basic fruit type was a follicle, a dry pod formed from a single carpel that splits along one seam to release seeds. Archaefructus carried exactly this kind of fruit. Many of the most ancient living angiosperm lineages still produce simple dry fruits or small drupes, reinforcing the idea that the ancestral fruit form was modest. Dehiscent fruits, those that pop or split open at maturity, rely on a zone of mechanical weakness in the carpel wall. In some species, that weakness forms along the dorsal vascular bundle of the carpel; in carpels that do not split open, that weakness zone simply never develops.8PubMed. Fruit dehiscence mechanism and release of dimorphic seeds with different germination properties in Commelina erecta This split-or-stay-shut mechanism is ancient and appears to have been fine-tuned many times over the course of angiosperm evolution.
Amborella trichopoda, a shrub found only in New Caledonia, sits at the very base of the living angiosperm family tree. It produces small, red, fleshy drupes. Because of its phylogenetic position, Amborella is often studied as a stand-in for what the earliest flowering plants might have been like, though researchers are careful to note that Amborella has had just as much time to evolve as any other living species, so its traits are not a frozen snapshot of the past.
Animals and the Rise of Fleshy Fruits
Fleshy fruits are essentially bribes. A plant wraps its seeds in nutritious, colorful tissue to attract an animal that will eat it and deposit the seeds somewhere new. This co-evolutionary bargain is so dominant today that it is easy to assume it has always been central to fruit evolution. The reality is more nuanced, but animal dispersal does appear to be ancient.
One of the most striking pieces of evidence comes from Early Cretaceous fossils of Jeholornis, a primitive bird from northeastern China. Preserved gut contents show that this bird was eating fruits long before the modern bird lineage even existed. Researchers have argued that early frugivorous birds may have expanded the scope for plant dispersal, potentially contributing to the rapid diversification of fruit-bearing angiosperms during the Cretaceous.9eLife. Earliest evidence for fruit consumption and potential seed dispersal by birds The idea is that mobile, flying seed dispersers could carry seeds far beyond what wind or gravity could achieve, opening up new habitats for colonization.
Even setting birds aside, the proportion of fleshy-fruited species among early angiosperms was higher than researchers once assumed. A study of Early Cretaceous plant assemblages from Portugal found that about a quarter of the angiosperm species present had fleshy fruits, a figure the authors described as surprisingly high for such an early stage in flowering plant evolution. They suggested that reptiles and early mammals (multituberculates, an extinct group of small, rodent-like creatures) were the likely seed dispersers.10PubMed. Seed Size and Dispersal Systems of Early Cretaceous Angiosperms from Famalicão, Portugal So the plant-animal fruit partnership was not a late development. It was woven into angiosperm ecology from fairly early on, even if the animals involved looked nothing like the monkeys and toucans that dominate the story today.
The Diversification Boom
For most of the Cretaceous, angiosperm fruits stayed small. But the trend of increasing seed and fruit sizes actually began before the asteroid impact that ended the Cretaceous 66 million years ago.7PubMed. Seed Size, Fruit Size, and Dispersal Systems in Angiosperms from the Early Cretaceous to the Late Tertiary After the mass extinction wiped out the non-avian dinosaurs and reshaped ecosystems worldwide, angiosperms exploded in diversity. The fraction of animal-dispersed species increased sharply in the early Paleogene (the period right after the Cretaceous), and fruit and seed diversity reached a peak during the Eocene, roughly 55 to 50 million years ago.11PubMed. Evolution of angiosperm seed disperser mutualisms: the timing of origins and their consequences for coevolutionary interactions between angiosperms and frugivores During that Eocene peak, new fruit types and new frugivore species were expanding into ecological niches that had not previously existed. It was a golden age for the fruit-animal partnership.
Then things contracted. By the late Eocene and Oligocene, global cooling set in, tropical forests shrank, and both seed and fruit sizes declined. Wind dispersal, which had been proportionally less important during the warm Eocene, bounced back.7PubMed. Seed Size, Fruit Size, and Dispersal Systems in Angiosperms from the Early Cretaceous to the Late Tertiary The pattern makes intuitive sense: fleshy fruits and the animals that eat them thrive in warm, forested environments. When forests give way to open grasslands or drier conditions, wind becomes a more reliable dispersal agent.
An interesting counterpoint to the idea that animal co-evolution was the main engine of fruit diversification comes from research looking at the deep Mesozoic record. Gymnosperms had already achieved a broad range of fleshy seed covering sizes by the Late Triassic, before angiosperms even appeared. As angiosperms diversified from small herbaceous ancestors, their fleshy fruits grew larger, but this did not dramatically change the overall size distribution of fleshy seed structures across all seed plants. The researchers behind this finding suggested that physical and life-history constraints, rather than co-evolution with frugivores, may be the primary force shaping how large fleshy fruits can get over deep evolutionary time.12PubMed. Patterns of variation in fleshy diaspore size and abundance from Late Triassic-Oligocene In other words, the basic rules of what is physically possible for a fleshy seed structure may matter more than the specific animals available to eat it.
The Genetic Toolkit Behind Fruit Formation
Across flowering plants, a shared family of genes called MADS-box transcription factors controls how flowers form and how carpels develop into fruits. In the classic model of flower development, different combinations of these genes specify the identity of sepals, petals, stamens, and carpels. When it comes to fruit development specifically, additional members of the MADS-box family are needed to ensure that the carpel matures properly after fertilization.13PubMed. Functional Genomics and Genetic Control of Flower and Fruit Development in Medicago truncatula: An Overview Tweaks to these genes can change the shape, size, and behavior of a fruit: whether it splits open or stays sealed, how thick the wall becomes, and even the overall geometry of the pod or berry.
What makes this relevant to the “first fruit” question is that these MADS-box genes are deeply conserved. The same gene families are present in gymnosperms, where they help build cones and fleshy seed coats. The evolutionary step to building a true fruit did not require inventing new genes from scratch. It required repurposing and tweaking an existing developmental toolkit in the context of a newly evolved structure, the carpel. The genes were ready. What changed was the anatomy they were working with.
Fruits in Water
While most of the fruit story played out on land, some angiosperms eventually carried their fruits into aquatic habitats. Transitions from land to water started almost as early as angiosperms themselves, but the pace accelerated between roughly 40 and 10 million years ago. Today, about 4,500 angiosperm species are aquatic, representing roughly one and a half percent of described angiosperm diversity.14PubMed Central. Macroevolutionary dynamics in the transition of angiosperms to aquatic environments Aquatic genera tend to be less species-rich than terrestrial ones, averaging around 11 species per genus compared to about 23 for land plants. Moving into water places severe constraints on fruit and seed dispersal strategies: wind dispersal becomes irrelevant, and the animals available to eat fruits underwater are a different cast entirely. Many aquatic angiosperms have shifted to water-dispersed or self-dispersed seeds, and their fruits often reflect those pressures with thinner walls, air-filled chambers for buoyancy, or mucilaginous coatings that help seeds stick to waterfowl.
Arils and Other Tricks
Not all seed-dispersal structures in angiosperms are technically parts of the fruit. Arils, fleshy outgrowths that develop around the seed itself rather than from the ovary wall, appear in many plant families and have sometimes been described as extra seed coats. They serve a similar ecological function to a fleshy fruit: they attract animals, provide a food reward, and get the seed moved somewhere new.15PubMed Central. Perspectives for a Framework to Understand Aril Initiation and Development The nutmeg’s bright red mace is an aril. So is the fleshy white tissue around a lychee seed. These structures blur the boundary between what is fruit and what is seed accessory, a reminder that evolution is not interested in tidy categories. The same dispersal problem, getting seeds away from the parent plant and into a good germination site, has been solved over and over with different anatomical raw materials, in gymnosperms, in ancient angiosperms, and in modern ones alike.