Multicellularity was supposed to be a threshold. Single-celled organisms crossed it; the crossing was irreversible; everything after was architecture. The history of animal life begins, in this framing, with a commitment: cells that once lived independently began dividing without separating, and the resulting clonal groups evolved into bodies. The alternative route — aggregation, where independent cells join together — was thought to be a different kind of thing entirely, the province of slime molds and social amoebae, not the lineage leading to animals.
The distinction mattered because it was supposed to be structural. Clonal multicellularity produces organisms where every cell is genetically identical, enabling the division of labor that makes complex bodies possible. Aggregative multicellularity produces chimeric groups, genetically heterogeneous, prone to conflict, limited in the complexity they can achieve. The classification organized an entire literature: animals are clonal, dictyostelids are aggregative, and the two paths lead to different evolutionary outcomes.
Ros-Rocher et al. studied Choanoeca flexa, a choanoflagellate that lives in ephemeral splash pools on the island of Curacao. Choanoflagellates are the closest living relatives of animals — they're the organisms we look at when we want to understand what the ancestors of animal multicellularity were doing. Choanoeca flexa forms motile sheets of cells: monolayers that swim, contract, and respond to light. The question was how these sheets form.
The answer is: all three ways. At low cell density, sheets form clonally — a single cell divides repeatedly. At high density, sheets form by aggregation — independent cells join together. At intermediate density, both happen simultaneously. The pathway isn't genetically fixed. It's environmentally tuned by salinity and cell density. When the splash pool evaporates and salinity rises past 73 parts per thousand, the sheets dissociate into individual cysts. When the rain refills the pool and salinity drops, sheets reform — by whichever route the local conditions favor.
The transitions are reversible. Choanoeca flexa goes unicellular, multicellular, unicellular again, with each cycle. Multicellularity is not a threshold it crosses but a state it enters and exits. The organism doesn't commit to being multicellular. It uses multicellularity when conditions call for it and discards it when they don't.
This doesn't just add a third category to the clonal-aggregative dichotomy. It dissolves the dichotomy. The same organism, with the same genome, uses clonal development when cells are sparse and aggregation when cells are abundant, depending on what's available. The deep assumption — that the route to multicellularity is a fixed property of a lineage — turns out to be false at the exact phylogenetic position that matters most for understanding animal origins.
The implication is that the ancestors of animals may not have committed to clonal multicellularity in one decisive evolutionary step. They may have been doing what Choanoeca flexa does: toggling between unicellular and multicellular states, using whatever combination of clonal and aggregative development the environment demanded. The commitment came later. The flexibility came first.
What interests me is the structure of the error. Clonal-versus-aggregative wasn't a lazy classification. It was supported by real biological differences — genetic homogeneity versus heterogeneity, complex versus simple body plans, different phylogenetic distributions. The categories did real work. They just also closed a door: if you know the answer is either A or B, you don't look for AB. You especially don't look for organisms that switch between A, B, and AB based on rainfall.
The finding is in one species. Whether it generalizes across choanoflagellates, or whether it reflects the ancestral condition, is uncertain. But the existence proof is enough to restructure the question. Multicellularity doesn't have to be a threshold. It can be a dial.