A carpel is the female reproductive organ of a flowering plant, the structure that houses and protects the ovules that, once fertilized, become seeds. Every fruit you have ever eaten developed from one or more carpels. It is, in a sense, the defining invention of flowering plants: before carpels evolved, seeds sat exposed on open, leaf-like structures. When the carpel appeared, enclosing those seeds inside a sealed chamber, it set the stage for the extraordinary diversification of the group that now dominates nearly every terrestrial ecosystem on Earth.
Anatomy of a Carpel
A carpel has three main regions that you can often see with the naked eye if you slice a flower open. At the base sits the ovary, a hollow chamber where ovules develop. Rising from the top of the ovary is the style, a slender stalk, and at the tip of the style is the stigma, the sticky or feathery surface where pollen grains land. Together, these three parts form a unit sometimes called a pistil, though the terminology gets a little tangled because a pistil can be made of one carpel or several fused together.
Inside the ovary, the walls have distinct inner and outer faces. The outer (abaxial) surface and the inner (adaxial) surface have different cell types, and this polarity runs all the way around the carpel’s circumference. One specialized tissue, the placenta, forms only on the inner side, right at the junction between the internal partition (the septum) and the ovary wall. The placenta is where ovules physically attach and receive nutrients, so its precise positioning matters enormously for fertility.
A Leaf That Became Something Else
Botanists have long proposed that carpels are modified leaves, and genetic experiments over the past few decades have confirmed that the idea holds up well. In the model plant Arabidopsis, when all the major genes that specify floral organs are knocked out, the flower’s parts revert to simple leaves. Conversely, forcing certain combinations of those identity genes to switch on in leaves can transform them into floral organs, including carpel-like structures.1Journal of Experimental Botany. An evolutionary perspective on the regulation of carpel development The carpel is not just any modified leaf, though. It is thought to be most directly related to ancient sporophylls, which were leaves that bore spore-producing organs in the ancestors of flowering plants. In those pre-angiosperm lineages, megasporophylls subtended ovules on their surfaces without enclosing them, much like the open cone scales you can see on a pine tree today.
The transition from an open, ovule-bearing leaf to a sealed carpel remains one of the great unsolved puzzles of plant evolution.2PubMed Central. Case not closed: the mystery of the origin of the carpel Researchers still debate exactly how the edges of the ancestral sporophyll folded inward and fused to create a closed chamber. Several competing models exist, and fossil evidence, while tantalizing, has not yet settled the question.
Why Enclosing Seeds Was a Game-Changer
Before carpels, seeds developed in the open, vulnerable to drying out, being eaten, or landing in the wrong conditions. Once seeds were enclosed, the parent plant gained a remarkable degree of control. The carpel could regulate which pollen reached the ovules, nourish the developing seeds, and eventually transform into a fruit that could lure animals, catch the wind, or float on water to carry seeds to new locations.3PubMed Central. Case not closed: the mystery of the origin of the carpel
There is a deeper evolutionary advantage to the closed carpel that goes beyond physical protection. A closed structure forces pollen tubes to grow through a narrow internal channel, which means the tubes have to compete with one another. That competition is essentially a fitness test: pollen from plants with poorly balanced genomes tends to produce slower-growing tubes that lose the race. The combination of insect pollination delivering diverse pollen and the closed carpel filtering it through tube competition may have functioned as a powerful screening mechanism, weeding out weaker genomes at very low cost to the parent plant. This hypothesis suggests that flowering plants could exploit the benefits of sexual recombination more efficiently than their open-seeded predecessors, helping them rise to ecological dominance.4PubMed. The rise of the angiosperms: a genecological factor
How a Flower Knows to Make a Carpel
Flowers develop their organs in concentric rings called whorls. The outermost whorl produces sepals, the next produces petals, then come the stamens (the male organs), and the innermost whorl produces carpels. A family of genes, broadly grouped into A, B, C, and E functions, determines which organs form in each whorl. Carpel identity depends heavily on the C-function gene known as AGAMOUS (AG). When AG is active alongside the E-function genes, the innermost whorl develops into carpels. When AG is broken, something strange happens: instead of carpels, the innermost whorl generates more petals and then starts repeating the flower pattern, producing a flower-within-a-flower that never forms reproductive organs.5The Plant Cell. Expression of the Arabidopsis floral homeotic gene AGAMOUS is restricted to specific cell types late in flower development
AG does not build the carpel single-handedly. It works by switching on a cascade of downstream genes, many of which are themselves regulators that manage specific aspects of carpel architecture. One of AG’s key roles is suppressing the leaf development program: without that suppression, the emerging floral tissue would default to making ordinary leaves rather than the specialized reproductive structure.6PubMed Central. Control of Reproductive Floral Organ Identity Specification in Arabidopsis by the C Function Regulator AGAMOUS This is consistent with the idea that the carpel is an evolutionary descendant of the leaf: genetically, the default state of the tissue is “leaf,” and making a carpel requires actively overriding that program.
The Internal Assembly Line
Once a carpel’s identity has been established, its internal tissues still need to differentiate. Two of the most important internal structures, the ovules and the septum that divides the ovary into chambers, arise from a specialized growth zone called the carpel margin meristem (CMM). In Arabidopsis, where two carpels fuse to form the gynoecium, the CMM runs along the margins where the two carpels meet. It generates the placenta, the ovules, the septum, the transmitting tract, and the stigma and style tissues.7PubMed. Inside the gynoecium: at the carpel margin Genes called CUC1 and CUC2 promote the formation and maintenance of this meristematic zone; when they fail, the internal organs develop poorly or not at all.8PubMed Central. The CUC1 and CUC2 genes promote carpel margin meristem formation during Arabidopsis gynoecium development
The transmitting tract deserves special attention. It is a column of highly secretory cells that runs from the stigma through the style and into the ovary, producing a rich extracellular matrix that pollen tubes need for nutrition, adhesion, and directional guidance as they grow toward the ovules.9PubMed Central. Paving the Way for Fertilization: The Role of the Transmitting Tract In mutants where the transmitting tract fails to form, pollen tubes stall or wander aimlessly, and fertility drops sharply. One well-studied example involves the gene NO TRANSMITTING TRACT (NTT): when NTT is nonfunctional, the mutant plants set only about 40% as many seeds as normal plants, and nearly all of those seeds cluster in the upper portion of the fruit because the pollen tubes cannot travel far without the tract’s support.10Current Biology. The Formation and Function of the Female Reproductive Tract in Flowering Plants
Fused Carpels and Their Advantages
Many familiar flowers have carpels that are fused together rather than separate. A tomato, for example, has several carpels joined into a single structure, and that fusion is called syncarpy. It is overwhelmingly common: the majority of flowering plant species have fused carpels to some degree. The question of why is surprisingly interesting.
When carpels fuse, pollen tubes can cross between them through shared transmitting tissue. That means a pollen grain landing on one part of the stigma can potentially fertilize ovules in a different carpel than the one it entered, which reduces the number of ovules that go unfertilized simply because pollen happened to land on the wrong side. A large evolutionary analysis found that roughly three-quarters of the transitions to syncarpy in flowering plant history resulted in pollen tubes gaining the ability to cross between carpels and reach ovules that would otherwise have been missed.11Journal of Evolutionary Biology. Evolution of syncarpy in angiosperms: theoretical and phylogenetic analyses of the effects of carpel fusion on offspring quantity and quality Most of those transitions also intensified pollen competition, reinforcing the fitness-screening mechanism already discussed. So syncarpy appears to boost both the number and the quality of seeds a plant produces.
Gatekeeping at the Stigma
The carpel does not passively accept all pollen. Many species have self-incompatibility systems that allow the carpel’s tissues to recognize and reject pollen from the same individual or close relatives, forcing outcrossing. Different plant families have evolved strikingly different molecular strategies for this. In the nightshade family, which includes tomatoes and tobacco, the pistil deploys a ribonuclease enzyme that destroys the RNA inside incompatible pollen tubes, effectively killing them. In the mustard family, a receptor on the stigma surface triggers a signaling cascade that blocks incompatible pollen from even germinating.12PubMed Central. Mechanisms of self-incompatibility in flowering plants Both systems rely on a highly diverse set of recognition genes, and the diversity itself is what makes the system work: because there are many versions circulating in a population, pollen is rejected only when it carries the same version as the pistil, which in practice means self-pollen or pollen from very close kin.
The Transition to Fruit
After successful pollination and fertilization, the carpel undergoes a dramatic transformation. The ovary wall thickens, cell division ramps up, and the structure begins to enlarge into what we recognize as a fruit. This transition is not automatic. It requires a hormonal trigger, primarily from gibberellin, a growth-promoting hormone whose production surges in response to fertilization. In tomatoes, mutant plants with a permanently activated gibberellin response will develop fruit even without pollination, which is how some seedless varieties are produced.13PubMed Central. Fruit setting rewires central metabolism via gibberellin cascades The gibberellin signal rewires the tissue’s metabolism, redirecting resources toward the expanding fruit. Without that signal, an unfertilized carpel typically withers and drops off the plant.
The sheer variety of fruits that result from this process is staggering. A peach is a single carpel that develops a fleshy outer layer and a hard inner stone. A raspberry is a cluster of many separate carpels from one flower, each becoming a tiny individual fruitlet. A pea pod is a single dried-out carpel that splits open to release seeds. An apple develops from fused carpels embedded in a fleshy receptacle. Every one of these forms traces back to the same basic organ.
What Mutants Have Revealed
Much of what we know about carpel development comes from studying plants with broken genes. In Arabidopsis, two mutants with evocative names have been particularly informative. SPATULA (SPT) mutants have carpels that fail to fuse at the top and lack a transmitting tract. CRABS CLAW (CRC) mutants produce shorter, wider carpels that also fail to close properly and lose their nectaries. When both mutations are combined, the defects are worse than you would expect from simply adding the two phenotypes together, which tells researchers that the two genes have overlapping roles in shaping the carpel.14Genetics. How flower development genes were identified using forward genetic screens in Arabidopsis thaliana
Perhaps more surprising is what happens in rice, which is also a flowering plant but belongs to a very distantly related lineage. In Arabidopsis, the C-function gene AGAMOUS is the master switch for carpel identity. In rice, knocking out both copies of the corresponding C-function genes (OsMADS3 and OsMADS58) does not eliminate carpels entirely. Carpel-like organs still form, driven by a different gene called DROOPING LEAF. This means the genetic wiring for carpel specification is not universal across flowering plants: rice uses a substantially different circuit than Arabidopsis does, which suggests that the genetic control of carpel identity has been rewired over evolutionary time.15Plant and Cell Physiology. Rice Flower Development Revisited: Regulation of Carpel Specification and Flower Meristem Determinacy
Polarity Genes That Work Across Species
While the identity genes can differ between lineages, some of the genes controlling carpel shape and symmetry are remarkably conserved. The YABBY family of genes, which help establish the inner-outer polarity of carpels, work in distantly related plants. In one experiment, a YABBY gene from wild sugarcane was transferred into an Arabidopsis mutant that had lost its own version of the gene (the CRC mutant, with those characteristic curling leaves and defective carpels). The sugarcane gene successfully rescued the mutant, restoring normal carpel closure and ovule development.16PubMed Central. Expression characterization and cross-species complementation uncover the functional conservation of YABBY genes for leaf abaxial polarity and carpel polarity establishment in Saccharum spontaneum Polarity establishment, in other words, uses older and more deeply shared genetic machinery than organ identity does. The decision about which tissue faces inward and which faces outward appears to predate the split between grasses and broadleaf plants.
Heat Stress and Carpel Failure
Carpel development is sensitive to environmental conditions, and high temperatures are a particular threat. In canola (Brassica napus), researchers found that a period of heat stress during flowering disrupted both the male and female reproductive tissues. When the stress was removed, plants did not immediately bounce back to full seed production, indicating that the damage to carpel development was not just about pollen failure on the male side. Disrupted carpels contributed directly to fruit abortion and reduced seed set.17Journal of Experimental Botany. High temperature stress of Brassica napus during flowering reduces micro‐ and megagametophyte fertility, induces fruit abortion, and disrupts seed production This finding matters for agriculture in a warming climate. Breeding programs that focus only on heat-tolerant pollen may miss half the problem if the carpel itself is also failing under stress.
Fossils and the Deep Past
Fossil evidence for early carpels is sparse and often ambiguous, but a few key discoveries have shaped how researchers think about carpel origins. One notable find from the Yixian Formation in northeastern China, dated to roughly 125 million years ago during the Early Cretaceous, preserves a fruit called Liaoningfructus. Its structure suggests the carpel that produced it was ascidiate, meaning it formed like a tube or sac rather than by folding a flat leaf inward along its edges. The fruit contained two seeds, with at least one receiving its vascular supply from the base, implying basal placentation. Both features challenge older textbook models that assumed the earliest carpels were flat structures that folded and sealed along a margin.18Acta Geologica Sinica – English Edition. The Earliest Ascidiate Carpel and Its Implications for Angiosperm Evolution
Fossils like Liaoningfructus do not settle the debate over carpel origins on their own, but they provide concrete constraints. They show that by 125 million years ago, carpels were already enclosed and already diverse in form, which pushes the origin of the structure even further back in time. Molecular clock estimates and fragmentary earlier fossils suggest that flowering plants, and therefore carpels, appeared sometime in the Late Jurassic or very early Cretaceous, but the picture remains incomplete. The transition from open sporophyll to closed carpel happened in a lineage that left frustratingly few fossils behind.