A leech does not have 32 brains, but the claim is rooted in real anatomy. What a leech actually has is 32 ganglia, which are clusters of nerve cells distributed along the length of its body like beads on a string. Each ganglion can independently process some sensory information and control local movements, which is probably how the “32 brains” myth took hold. The reality is more interesting than the headline: leeches have a nervous system that blurs the line between a centralized brain and a fully distributed network, and that architecture has made them one of neuroscience’s most useful research animals.
What the 32 Ganglia Actually Are
The leech central nervous system sits inside a blood-filled sinus that runs the length of the animal. It consists of 32 segmented ganglia connected to one another by nerve fiber bundles called connectives.1Molecular Immunology. Evidence for a novel chemotactic C1q domain-containing factor in the leech nerve cord Each ganglion contains a few hundred neurons, a tiny number compared to the roughly 86 billion in a human brain. But those neurons are large and individually identifiable under a microscope, which means researchers can map them, name them, and study the same specific neuron across different animals.
Not all 32 ganglia are equal. The ones in the middle body segments are structurally similar to one another, each serving a roughly equivalent stretch of body wall and muscle. But at both ends of the animal, ganglia have fused together into larger, more complex structures that take on coordinating roles more like what we would recognize as a brain.
The Head Brain and What It Controls
At the front end, several ganglia have merged into a compound structure sometimes called the “head brain.” This consists of two main parts: the supraesophageal ganglion, which sits above the esophagus, and the subesophageal ganglion, which sits below it. Together, they process input from the leech’s eyes and other sensory organs on its head, and they exert top-down control over the body.
One of the clearest demonstrations of what the head brain does comes from removal experiments. When the entire head brain is taken out, a leech loses the ability to perform coordinated crawling. But swimming actually gets easier to trigger and lasts longer.2PubMed Central. Specialized brain regions and sensory inputs that control locomotion in leeches That seems contradictory until you realize the two halves of the head brain play different roles. When only the supraesophageal ganglion is removed but the subesophageal ganglion is left intact, swim episodes become dramatically longer, with some preparations swimming virtually continuously. In contrast, preparations that retained the full head brain swam for much shorter durations.2PubMed Central. Specialized brain regions and sensory inputs that control locomotion in leeches
What this tells us is that the supraesophageal ganglion acts partly as a brake on swimming, while the subesophageal ganglion helps drive it. The head brain is not just a passive processing center; different regions within it push behavior in opposing directions, creating a kind of internal negotiation about what the animal should be doing at any given moment.
The Tail Brain and Its Hidden Complexity
The rear end of the leech has its own compound structure: the caudal ganglion, which provides sensory and motor control to the posterior sucker the animal uses for attachment. This ganglion is actually the product of seven embryonic ganglia that fused during development. Despite the fusion, each of those seven contributing units, sometimes called subganglia, can still be distinguished both structurally and functionally.3PubMed. The caudal ganglion of the leech, with particular reference to homologues of segmental touch receptors
The touch-sensitive cells in the caudal ganglion are organized in a pattern that reflects the modified segmentation of the round sucker. Their receptive fields overlap more extensively than those of equivalent cells in the mid-body ganglia, giving the sucker finer and more redundant sensory coverage.3PubMed. The caudal ganglion of the leech, with particular reference to homologues of segmental touch receptors That makes sense for a structure that needs to grip surfaces reliably: extra overlap means the sucker can detect and respond to contact even if some sensory neurons are damaged or obstructed.
How Independent Are the Ganglia?
This is the question at the heart of the “32 brains” claim. If each ganglion could think and act entirely on its own, calling them individual brains would be more defensible. The truth is somewhere in the middle. The leech nervous system is genuinely distributed: many behaviors can be controlled at the local level, within a single ganglion, much as vertebrate stretch reflexes are handled by one or a few spinal cord segments with modulation from higher centers.4Journal of Experimental Biology. Functions of the subesophageal ganglion in the medicinal leech revealed by ablation of neuromeres in embryos
A good example is the local bending reflex. When a sensory cell in one ganglion detects pressure on the body wall, motor neurons in that same ganglion can generate a bending response without waiting for instructions from the head brain. Researchers have shown that each pattern of sensory stimulation produces a unique pattern of motor neuron output, and the responses match what the whole animal would do behaviorally.5PubMed Central. Distributed processing of sensory information in the leech. I. Input-output relations of the local bending reflex In other words, a single ganglion can sense what is happening to its segment, decide on an appropriate muscle response, and execute it. That is a remarkable degree of autonomy for a structure containing a few hundred neurons.
But autonomy is not the same as independence. For whole-body actions like swimming or crawling, the ganglia need to coordinate with one another, and they depend on the head brain for decisions about when to start or stop. A ganglion handles its own local affairs; the head brain and the connections between ganglia handle the big picture.
Coordinating a Swimming Wave
Leech swimming is one of the best-understood rhythmic behaviors in all of neuroscience. The animal undulates its body in a wave that travels from head to tail, and that wave depends on precise timing between segments. Each body ganglion contains its own pattern-generating circuit that can produce rhythmic output, but these circuits must be linked so that each fires slightly after the one ahead of it.
The traveling wave sweeps along the body with a roughly constant phase lag between segments, around 20 degrees per segment during normal swimming. This ensures that about one full wavelength is expressed along the body during each cycle, which lasts anywhere from about 0.4 to 2.0 seconds.6Seminars in Neuroscience. Mechanisms of intersegmental coordination in leech locomotion The coordination is achieved by interneurons that project between ganglia in both directions, creating a chain of coupled oscillators. Computer models of this system can reproduce the phase relationships seen in real nerve cords by incorporating the known signal conduction delays and the fact that ganglia toward the tail tend to have slightly faster intrinsic rhythms.7PubMed. Model for intersegmental coordination of leech swimming: central and sensory mechanisms
This is a good illustration of why calling each ganglion a “brain” is misleading. Each one can oscillate on its own, yes. But the behavior that emerges, a smooth swimming wave, only works because the ganglia are coupled into a coordinated chain. They are more like the members of a drumline who each keep their own beat but are trained to stay slightly offset from the person next to them.
Serotonin, Hunger, and Decision-Making
Leeches also offer a surprisingly clear window into how a simple nervous system handles something that looks like a decision: whether to keep feeding or stop. The neurochemical serotonin plays a central role. Warming the leech’s lip is enough to trigger ingestion, and this stimulus excites serotonin-producing neurons into high-frequency firing. When the body wall stretches from a full blood meal, ingestion stops, and those same serotonin neurons are inhibited. Applying serotonin externally produces hyperphagic (overeating) behavior, while destroying serotonin-producing cells makes the leech behave as though it is already full.8PubMed. Serotonergic modulation of the feeding behavior of the medicinal leech
The serotonin signal is not confined to one ganglion. Measurements of serotonin levels across the full nerve cord show that recently fed leeches have about 28% less serotonin than hungry ones, and this depletion is consistent across the head brain and all segmental ganglia.9PubMed. Serotonin in the leech central nervous system: anatomical correlates and behavioral effects Even six weeks after feeding, leeches whose bodies remained distended with their blood meal had lower serotonin levels than those whose meals had been experimentally removed, suggesting that ongoing body-wall stretch suppresses serotonin production over the long term.9PubMed. Serotonin in the leech central nervous system: anatomical correlates and behavioral effects The feeding “decision” is not localized to one brain region. It is a whole-body signal mediated by a single chemical broadcast across the entire nervous system.
Can Leeches Learn?
They can, within limits. The medicinal leech shows habituation, the simplest form of learning: if you repeatedly tap a leech in the same spot, its shortening reflex gets weaker each time. Move the stimulus to a new location, and the degree to which the habituation carries over depends on how close the new spot is to the trained one. Closer spots show more generalization, partly because neighboring ganglia share some sensory cells.10PubMed Central. Generalization of habituation and intrinsic sensitization in the leech
Leeches also show sensitization, the opposite of habituation: a strong or noxious stimulus can temporarily amplify responses. Swimming evoked by weak electrical stimulation decreases with repetition in a pattern consistent with habituation, including spontaneous recovery after rest and faster decline in repeated training sessions. A brush across the skin can restore the diminished response, a phenomenon called dishabituation.11PubMed. Nonassociative learning in the leech Hirudo medicinalis These are nonassociative forms of learning, meaning the leech is not forming associations between events (as in Pavlovian conditioning) but is instead adjusting its responsiveness based on recent experience. It is simple, but it is real learning, and it emerges from the properties of individual ganglia and the connections between them.
Why Neuroscientists Keep Coming Back to Leeches
The medicinal leech has been a standard model in neuroscience for decades, and the reason is largely architectural. Its neurons are big enough to see, few enough to map, and consistent enough from animal to animal that a researcher can locate the same identifiable cell in every specimen.12Journal of Experimental Biology. A classic model animal in the 21st century: recent lessons from the leech nervous system That has made the leech central to studies that connect the properties of single neurons and their wiring to the functioning of the whole animal.
One especially striking demonstration involved placing a single identified leech neuron, a Retzius cell, onto the gate of a silicon transistor. The neuron’s action potentials directly modulated the electrical current flowing through the transistor, creating a biological-electronic hybrid at the single-cell level.13PubMed. A neuron-silicon junction: a Retzius cell of the leech on an insulated-gate field-effect transistor Experiments like this are possible because leech neurons are robust, well-characterized, and large enough to manipulate individually. The same cell type can also be removed from a ganglion and kept alive in culture, where it will form functional chemical synapses with other identified neurons.14PubMed Central. Chemical transmission between individual Retzius and sensory neurones of the leech in culture
The leech’s neurochemical toolkit also turns out to be richer than you might expect from an animal with such a small nervous system. Transcriptomic studies of the leech CNS have identified receptors for GABA, glutamate, and acetylcholine in numbers and varieties that parallel those found in much more complex animals. The acetylcholine receptor family alone includes around 20 distinct receptor types, split between muscarinic and nicotinic subtypes.15PLOS ONE. An annotated CNS transcriptome of the medicinal leech, Hirudo verbana: De novo sequencing to characterize genes associated with nervous system activity That kind of molecular diversity in a nervous system with only a few thousand total neurons gives researchers a chance to study how neurotransmitter systems work without the overwhelming complexity of a mammalian brain.
A Nervous System That Can Repair Itself
Unlike the mammalian central nervous system, the leech CNS can regenerate functional connections after injury. Cut a connective between two ganglia, and the axons will sprout and eventually restore synaptic connections that bring back normal behavior.16PubMed Central. The leech nervous system: a valuable model to study the microglia involvement in regenerative processes This regenerative ability depends on immune-like cells called microglia that accumulate at the injury site. When microglial accumulation is experimentally reduced, axon sprouting drops significantly.17PubMed. Reduced axon sprouting after treatment that diminishes microglia accumulation at lesions in the leech CNS
Even more unexpectedly, introducing a controlled bacterial infection at the injury site enhances the regenerative process. The immune response triggered by microbial challenge leads to the production of antimicrobial peptides by both microglial cells and neurons themselves. These peptides do double duty: they fight bacteria and they promote axon regrowth.18PubMed Central. Microbial challenge promotes the regenerative process of the injured central nervous system of the medicinal leech by inducing the synthesis of antimicrobial peptides in neurons and microglia Understanding why the leech can do something our nervous system largely cannot is one of the active reasons researchers continue to study it. The answers may eventually inform strategies for promoting neural repair in mammals.
Segmented Nervous Systems in Evolutionary Context
The leech’s chain-of-ganglia design is not unique. Many invertebrates with segmented bodies, including earthworms and insects, have nervous systems built on a similar plan: a series of ganglia, one per segment, connected by nerve cords. For a long time, biologists assumed this “rope ladder” nervous system was inherited from a common ancestor shared with arthropods. Phylogenetic analysis has complicated that story. When the distribution of segmented nervous systems is mapped across the tree of life, the pattern does not fit neatly with a single ancestral origin. The rope-ladder layout in annelids like leeches and in arthropods like insects may represent convergent evolution rather than direct inheritance from a shared ancestor.19PubMed Central. Embracing the comparative approach: how robust phylogenies and broader developmental sampling impacts the understanding of nervous system evolution
If that is the case, segmented ganglia may be something evolution has landed on more than once as a good solution to the engineering problem of controlling an elongated, flexible body. Each segment gets its own local processor, with just enough communication between segments to coordinate whole-body actions. It is a design philosophy more akin to a distributed computing network than to a single centralized computer, and it works well enough that evolution appears to have converged on it independently in lineages separated by hundreds of millions of years.
How Leech Neurons Respond to Anesthetics
Because individual leech neurons can be identified and recorded from reliably, they have also been useful for studying how drugs act on nerve cells. Procaine, a local anesthetic related to the novocaine used in dentistry, has different effects on different identified pain-sensing neurons within the same ganglion. In one type of nociceptive cell, procaine causes depolarization (making the cell more excitable), while in an adjacent type it causes hyperpolarization (making the cell less excitable). Both effects are accompanied by a decrease in the cell’s input resistance and are reversible after the drug is washed away. The depolarization depends on sodium ions and disappears in sodium-free solutions, while the hyperpolarization persists even when chloride levels are altered and is enhanced when potassium is reduced.20PubMed. Actions of procaine on specific nociceptive cells in leech central nervous system
The fact that a single drug can produce opposite effects on two neighboring neurons of the same functional class highlights something important about how nervous systems work: even cells with similar jobs can have different molecular machinery under the hood. In a mammalian brain with billions of neurons, that kind of cell-by-cell dissection is extraordinarily difficult. In a leech ganglion with a few hundred identifiable cells, it becomes tractable. That is the enduring appeal of the leech for basic neuroscience: not that it has 32 brains, but that its 32 ganglia are simple enough to understand in detail while still being complex enough to generate real behavior.