A pot of water sits on the stove. The temperature drops: plus one degree, zero, minus one — the water stays liquid. But at some point, a light breath is enough to make the whole mass instantly freeze into ice. Physicists call such jumps first-order phase transitions. Similar switches happen in living communities — only the role of temperature is played by other forces.
In the new work, scientists rigorously showed that microbial populations undergo genuine first-order phase transitions. As an example, they took bacteria carrying plasmids — small circular DNA molecules that float inside the cell separately from the main chromosome. A plasmid can confer antibiotic resistance or the ability to feed on a new substance, but copying and maintaining it costs the bacterium resources. If the benefit doesn't outweigh the cost, the plasmid becomes a burden.
The authors derived a strict lower bound: if the average number of plasmids per cell drops below a certain threshold, the entire population loses them in a jump. Above the threshold, plasmids persist; below it, they vanish almost instantly. No smooth transition. Such an abrupt breakdown explains why experiments with microbes often show an "all-or-nothing" pattern: either everyone has the trait, or no one does.
The link to cell lineages adds depth. Each bacterium divides, and its descendants inherit the plasmids. If, due to random fluctuations, a branch's plasmid count falls below the threshold, that branch is doomed, and soon the whole population follows. Thus individual histories combine into a collective tipping point. The practical takeaway is simple: biotechnologists who insert useful genes into bacteria can calculate in advance how many copies to maintain so that the culture doesn't "freeze" into an undesirable state.
First-order phase transitions are familiar to everyone: melting ice, boiling water, hardening chocolate. Now the same mathematical machinery describes the fate of an entire bacterial population.
In "Jurassic Park," the heroes filled gaps in dinosaur DNA with frog genes, and got unexpected jumps in behavior. Here, the jump arises not from foreign genes but from a simple count of their number.