1879: An Impossible Find Crawling Out of a Sponge
Imagine being William Carmichael M'Intosh in 1879—a Scottish physician and marine zoologist sampling waters near Cebu, the Philippines, aboard HMS Challenger as part of the Challenger Expedition. The ship hauls up a glass sponge, *Crateromorpha meyeri*, from 175 meters down. You dissect it and find a worm inside its labyrinthine canals.
Not an ordinary worm. It has one head, but its body branches repeatedly—one becomes two, two become four, four become eight—until it ends in a tangled mass of tails spread through every corner of the sponge.
M'Intosh wrote in his paper, with evident puzzlement, that the worm's body seemed to have a "furor" for budding—sprouting new branches laterally, terminally, at every broken point.
This was humanity's first encounter with a branching worm, *Syllis ramosa*. In the 140-plus years since, among the more than 20,000 known annelid species (including all earthworms, leeches, and ragworms), only three branching species have been found.
2022: King Ghidorah Arrives
In January 2022, Maria Teresa Aguado's team at the University of Göttingen published the third branching worm in *Organisms Diversity & Evolution*. The specimens came from shallow waters (0–20 m) off Sado Island, Japan, dwelling inside the canals of a *Petrosia* sponge.
They named it *Ramisyllis kingghidorahi*.
King Ghidorah—Godzilla's archenemy, a three-headed, two-tailed monster capable of regenerating its severed ends.
Aguado explained the naming: "King Ghidorah is a fictional branching creature that can regenerate lost ends, so we felt the name was fitting for this new branching worm."
The name sounds like a joke, but the biology is serious. The worm shares two real traits with its namesake: a branching body and regenerating ends. The difference: King Ghidorah has three heads, while *Ramisyllis kingghidorahi* has just one.
One head. Hundreds of tails.
A Tree Living Inside Sponge Canals
To grasp how strange this worm is, you first have to understand its home.
Sponges are not plants—they are animals, and among the oldest multicellular ones. Their bodies are riddled with canal systems: water flows in through pores, passes filter-feeding cells, and exits elsewhere. The canals branch, wind, and interconnect like a maze.
*R. kingghidorahi* lives inside that maze. Its head is buried in the sponge's center, feeding on organic particles carried by the water current. But its body is not a straight line—it branches like a tree. The trunk gives off side branches, which branch again, with each branch extending into one of the sponge's canals. Every branch has a complete set of organs: gut, nerves, muscle. All of them connect back to the single head.
One head feeding hundreds of tails. One mouth sustaining hundreds of bodies.
This is not parasitism—the worm does not drain the sponge's nutrients. Nor is it simple symbiosis—scientists still don't know what the sponge gets out of it. As Aguado put it: "We still don't understand the nature of the relationship between branching worms and their host sponges. Is it mutualism? How does the worm sustain its enormous body with one tiny mouth?"
May 2025: The First Gene Activity Map
In May 2025, the Göttingen team published a new paper in *BMC Genomics*—the first complete transcriptome of *R. kingghidorahi*.
What is a transcriptome? If the genome is the "blueprint," the transcriptome is the "construction plan"—it tells you which genes are actually activated, in which body parts, at which times. Every cell in an organism carries the same genes, but eye cells activate visual genes and muscle cells activate motor genes. The transcriptome is the map of "who is working where."
The team analyzed three body regions: the head, the mid-body, and the reproductive stolons. They compared males, females, and immature individuals to see which genes were active where.
The results were surprising.
Surprise 1: The head is not the control center
Scientists had assumed the head would house a sex-control center—a "command headquarters" deciding whether the whole animal becomes male or female. The data said no. Gene activity differences between male and female heads were far smaller than expected.
The real hotspot of gene activity was the stolon.
Surprise 2: The stolon is the true "individual"
What is a stolon? This is the weirdest part of the worm.
Come breeding season, the branch tips of *R. kingghidorahi* begin to swell, growing an independent reproductive unit—a stolon. The stolon has its own eyes, its own brain, its own muscles. It detaches from the parent, swims into the open sea, finds a stolon of the opposite sex to mate with, and then dies.
In other words, every branch tip of this worm can grow a "temporary individual"—with eyes, able to see, swim, and mate, but with a lifespan of only days.
The 2025 study found that eye-development genes are significantly upregulated in stolons. This is the first molecular evidence explaining how "branch tips become independent individuals."
Stolons are about 1.5 mm long. A worm with hundreds of branch tips can produce hundreds of stolons at once—hundreds of eye-bearing temporary individuals detaching from one tree and swimming out to sea.
Surprise 3: The mid-body is a "passive pipeline"
The head is active, the stolons are active—but the mid-body, the bulk of those hundreds of branches, shows remarkably flat gene activity. It seems to be a passive connecting pipeline, shipping food digested by the head to the tips, and relaying reproductive signals from the tips back to the head.
There is one strange exception: female mid-body gene activity is markedly higher than in males. Why? Nobody knows yet.
Surprise 4: Possible genome duplication
The data hint that *R. kingghidorahi* underwent partial genome duplication over its evolutionary history. Part of the genome was copied, giving some genes two copies—one keeps the original job, the other is free to evolve new functions.
This is not rare in biology—human ancestors went through two rounds of whole-genome duplication. But in annelids the phenomenon has barely been studied. If confirmed, genome duplication could explain how the worm evolved such a complex body plan—the extra gene copies gave it room to "experiment."
How Does One Head Decide Which Branches Reproduce?
This is the biggest mystery.
*R. kingghidorahi* has hundreds of branch tips, each capable of growing a stolon. But they do not all reproduce simultaneously. Some branches sprout stolons at a given moment; others stay dormant.
How does one head decide which branch reproduces and which waits?
The 2025 paper offers no answer. The data hint that "different branches may receive different signals, or respond at different times," but the triggering mechanism is entirely unknown.
The question is hard because it challenges our basic understanding of "individuality." Humans have a unified hormonal system—the brain issues orders, hormones travel through the blood, and all cells receive instructions at once. But *R. kingghidorahi*'s body is branching, dispersed, reaching into every canal of the sponge. How does a signal from one head reach hundreds of tips? At equal strength? Or does each branch have its own "local clock," triggering only when conditions are right?
The Boundaries of Individuality
Here is an angle I find fascinating.
We are used to thinking of "one organism" as a clear unit—one cat is one cat, one fish is one fish. But *R. kingghidorahi* blurs that line.
- Is it one individual? Yes—it has one head, one mouth, one digestive system.
- Is it a tree? Also yes—its body branches like a tree, each branch with its own organs.
- Are its stolons its children? Genetically, the stolons are part of itself (asexual reproduction). But each stolon has its own eyes, its own brain, its own behavior—it swims, sees, mates. That looks more like an independent individual.
- Aguado et al. (2022). *Ramisyllis kingghidorahi* n. sp., a new branching annelid from Japan. *Organisms Diversity & Evolution*. doi: 10.1007/s13127-021-00538-4
- Ponz-Segrelles et al. (2025). Sex-specific differential gene expression during stolonization in the branching syllid *Ramisyllis kingghidorahi*. *BMC Genomics* 26(1). doi: 10.1186/s12864-025-11587-w
Strangest of all: the stolon's eyes grow on the parent's branch tips. In other words, one "individual's" eyes grow on another "individual's" body. We don't know whether a stolon retains memories or can learn after detaching—it has a brain, but that brain grew only days ago.
This reminds me of an AI analogy. If you have a central model (the head) and hundreds of distributed agents (branches), each agent capable of independent perception, decision-making, and action, but all connected back to the central model—is that one system or hundreds? If an agent is dispatched on a mission (a stolon), carrying its own perception (eyes) and decision-making (brain), but its "identity" derives from the central model—is it a new individual, or a part of the parent?
Designers of distributed systems ask this constantly. *R. kingghidorahi*, after 500 million years of annelid evolution, offers an answer: it can be both. It is one individual (one head, one mouth) and a swarm of individuals (hundreds of eye-bearing stolons). The boundary of individuality is not a line but a spectrum.
Why Did Evolution Only Try This Three Times?
More than 20,000 annelid species, only 3 that branch. Not because branching is bad—it lets *Ramisyllis* perfectly exploit every inch of sponge canal space. But evolution rarely takes this path.
The reason may be that a branching body requires a "furious" combination: the lifelong ability to produce new segments (many worms have this) + strong regeneration + the ability to produce multiple new segments simultaneously. Each condition alone is not rare; together they almost never occur.
In other words, branching is not something evolution achieves in "one step." It requires three existing abilities pushed to their extreme in a single species. Evolution doesn't invent new tools—it pushes existing tools to their limits.
The Stranger Deep in the Sponge
In 1879, when M'Intosh first saw *Syllis ramosa* among the Challenger's specimens, he wrote of a "furor for budding." What a word. Not "ability," not "mechanism"—"furor." An organism with an almost obsessive insistence on where its body grows, how much, and when to stop.
146 years later, we have gene activity maps, transcriptomes, and molecular-level explanations. But that original puzzlement remains: one head, hundreds of tails, hundreds of eye-bearing temporary individuals detaching and swimming away—is this one living being?
Maybe. Maybe the concept of "one" simply doesn't apply inside a sponge's canals.
---
References: