The Blob
AmysteriousorganismintheParisZoohas720sexes,nobrain,nostomach,noeyes,yetitcanlearn,solvemazes,andteachotherblobswhatitlearned.Cutitinhalfandithealsintwominutes.It'snotaplant,animal,orfungus.It'sbeenaliveforabillionyears.
What Is This Thing?
The first problem with the blob is that nobody knows what to call it.
For centuries, Physarum polycephalum was classified as a fungus. It looks fungal — it grows on decaying matter in forests, it spreads in branching networks, it produces spore-bearing fruiting bodies. But it moves. It crawls across surfaces at speeds up to 4 centimetres per hour, flowing like a slow-motion amoeba. Fungi don't do that.
It was reclassified as a protist — a catch-all kingdom for organisms that don't fit neatly into animal, plant, or fungus categories. More recently, it's been placed in the supergroup Amoebozoa, within the class Mycetozoa (slime moulds). Its lineage diverged from other eukaryotes roughly one billion years ago.
It is a plasmodium — a single cell containing millions of nuclei. Unlike most organisms, which are made of trillions of individual cells, the blob is one continuous cell. The nuclei divide, but no cell walls form between them. The entire organism is a single, vast, multinucleated cytoplasm, enclosed in one continuous membrane.
It is the largest single cell you will ever see. One specimen at the Paris Zoo covered 10 square metres. One cell. No brain. No nervous system. And it can solve problems that stump some algorithms.
How It Moves
The blob moves by cytoplasmic streaming — rhythmic contractions that push its internal fluid (cytoplasm) in one direction, extending the cell membrane forward and retracting it behind. Think of it as a very slow, very deliberate ooze.
But the movement isn't random. The blob makes decisions about where to go.
Place a blob on a surface with multiple food sources (it eats bacteria, fungal spores, and oat flakes in lab settings). The blob will initially spread outward in all directions, exploring. As it discovers food, it redirects its flow toward nutrient-rich areas and withdraws from nutrient-poor ones.
Within hours, the sprawling exploration network prunes itself down to an efficient transport system connecting all food sources via the shortest possible routes. The blob has solved an optimisation problem — without a brain, without a plan, without any centralised control.
The Tokyo Rail Experiment
In 2010, a team of researchers at Hokkaido University conducted one of the most famous experiments in modern biology.
They placed a Physarum blob in the centre of a map of the Greater Tokyo area. At positions corresponding to the major cities and towns around Tokyo, they placed oat flakes. Then they waited.
The blob expanded outward, found the oat flakes, and over 26 hours, pruned its network down to a set of connections between the food sources.
The blob's final network was nearly identical to the actual Tokyo rail system — a network that human engineers had spent decades and billions of dollars designing. The blob did it in 26 hours with no brain and no blueprint.
The paper, published in Science, demonstrated that the blob's network was not only efficient but also fault-tolerant — it had redundant pathways that could compensate for breaks, similar to the actual rail system's design.
This wasn't a fluke. Subsequent experiments replicated the result with maps of the UK motorway system, the Roman road network, and the US interstate system. In each case, the blob produced networks comparable to those designed by human engineers.
Learning Without a Brain
In 2016, a team led by Audrey Dussutour at the Centre de Recherches sur la Cognition Animale in Toulouse published a landmark paper in the Proceedings of the Royal Society B.
They presented blobs with a bridge coated in quinine (bitter) or caffeine (unpleasant) that they had to cross to reach food. Initially, the blobs avoided the bridges. But over several days, they learned that the substances were harmless and began crossing without hesitation.
This was habituation — the simplest form of learning. The blob learned to ignore a noxious stimulus. A cell with no neurons had formed something analogous to a memory.
Then Dussutour did something extraordinary. She took a blob that had learned to cross the quinine bridge and fused it with a "naive" blob that had never encountered quinine. The fused blob crossed the bridge without hesitation. The learned behaviour had been transferred.
The mechanism is still debated. One theory is that the "memory" is encoded in the blob's cytoplasmic flow patterns — essentially, the physical structure of the cell retains information. When two blobs fuse, their cytoplasm mixes, and the flow patterns of the "educated" blob overwrite those of the naive one.
720 Sexes
Most organisms have two mating types (sexes). Physarum polycephalum has 720.
Each blob has a genetically determined mating type. Reproduction (via spores) requires the fusion of two blobs with different mating types. With 720 options, the odds of any two random blobs being compatible are approximately 99.86%.
This extreme diversity of mating types is thought to be an evolutionary strategy for maximising genetic diversity. In a population of blobs, virtually any encounter between two individuals can produce offspring. There is no "wrong" partner — only one in 720.
The concept of 720 sexes isn't quite the same as 720 genders in the human sense. Each "sex" in Physarum is a genetically distinct mating type. Any two blobs of different types can fuse and reproduce. Think of it as 720 different keys, where any two different keys can open the same lock.
Why It Matters
The blob is not just a curiosity. It's a philosophical challenge.
Intelligence — the ability to learn, decide, and adapt — is usually assumed to require a nervous system. Neurons, synapses, a brain. The blob has none of these. It is a single cell, operating on chemical gradients and physical forces, and yet it outperforms some computer algorithms at network optimisation.
This suggests that intelligence is not a product of neurons. It is a property of life itself — a capacity that evolution discovered long before it invented brains. The blob has been solving problems for a billion years. Our brains have been solving problems for about 500 million. We're the newcomers.
The Tokyo rail experiment
A blob recreated the Tokyo rail network in 26 hours. This video shows the time-lapse — watching a brainless cell solve an engineering problem that took humans decades.
Solving mazes is impressive. But can a cell without neurons actually learn?
The Dussutour learning experiments
Audrey Dussutour proved that blobs can learn — and then transfer their memories to other blobs by fusing with them. A cell without neurons is teaching.
720 sexes. Not a typo. Here's what that means.
The 720 mating types
Why does a single-celled organism need 720 different sexes? The answer tells you something profound about how evolution solves the genetic diversity problem.
What you now know
- Physarum polycephalum is a plasmodium — a single cell with millions of nuclei but no cell walls, capable of growing to cover 10 square metres
- In the famous 2010 Tokyo rail experiment, the blob recreated the rail network in 26 hours — matching a system that human engineers spent decades designing
- Dussutour's 2016 research proved the blob can learn to ignore noxious substances, and can transfer that learned behaviour to another blob through cell fusion
- The blob has 720 mating types, giving any two random blobs a 99.86% chance of being compatible — an evolutionary strategy for maximising genetic diversity