Here is the simplest living experiment anyone has ever run. Take one bacterium. Feed it. Every hour, it splits into two — each of those splits again the next hour, and so on. Nothing exotic, nothing engineered: E. coli in a nutrient broth doubles roughly every 20 minutes; many common lab strains settle to a once-per-hour doubling. It is not a metaphor. It is a measured fact.
The question is: how long before the petri dish is full?
Before reading on, write down your honest guess. Most people say a few days. A week. A month.
The actual answer is much, much shorter.
The clock and the counter
Drag the slider below from t = 0 hours to t = 30 hours. At every hour t, the bacterial population is N(t) = 2^{t} — starting from a single cell at t = 0. A horizontal line marks the carrying capacity: about 10^{9} cells, roughly the number of bacteria that fit in a standard 10-centimetre petri dish before the dish runs out of food and space.
The slope of the diagonal is what tells the story: on a \log_{2} axis, every doubling moves one tick up. The growth is straight, but it is straight in log space — which means in real space it is explosive.
Reading the numbers hour by hour
Let's walk the clock forward and see what each hour brings. Remember: start with one cell at t = 0.
- t = 0 h. N = 1. A single invisible speck.
- t = 1 h. N = 2. Still invisible.
- t = 10 h. N = 2^{10} = 1{,}024. About a thousand cells. Invisible still, but detectable by a lab instrument.
- t = 20 h. N = 2^{20} \approx 1.05 \times 10^{6}. A million cells. A faintly cloudy broth — the classic lab signal that something is growing.
- t = 24 h. N = 2^{24} \approx 1.7 \times 10^{7}. Seventeen million. Clearly cloudy.
- t = 30 h. N = 2^{30} \approx 10^{9}. A billion cells — the dish is saturated, the broth is milky, and the population has hit the carrying capacity. Growth slows or stops because there is no more food.
Why 2^{30} \approx 10^{9}: a handy conversion between base-2 and base-10 logs is 2^{10} = 1024 \approx 10^{3}. So 2^{30} = (2^{10})^{3} \approx (10^{3})^{3} = 10^{9}. Every ten doublings multiplies by roughly a thousand.
From one cell to a billion in just 30 hours. Not months. Not weeks. About a day and a quarter.
If E. coli doubled every 20 minutes — which is its rate in nutrient-rich conditions — the same billion-cell saturation would be reached in just 10 hours.
The carrying-capacity line
Why does the growth stop? The exponential curve wants to keep climbing. Real bacteria cannot.
A petri dish holds a finite amount of nutrient. When the bacteria eat it all, they stop dividing. The carrying capacity K is the maximum population the environment can sustain — for a petri dish, roughly 10^{9} cells per millilitre. On the plot, K is the dashed horizontal line at \log_{2}(K) \approx 30.
Before the curve hits that line, the growth is exponential — N(t) = 2^{t}, a pure doubling. After the curve reaches the line, the real biology takes over: cells stop dividing, some die, and the population plateaus near K. Biologists model this with the logistic curve, which is exponential at first and levels off near K. But the exponential phase — the part before saturation — is where our story lives, because it is the part that blindsides intuition.
Why the king-and-rice intuition fails
Two facts about exponential growth feel paradoxical the first time you meet them:
Fact 1. The last doubling wipes out half the history. If the dish is full at t = 30 hours, then at t = 29 it was half full. One hour earlier — when most observers would still have said "plenty of room" — the whole dish was one doubling away from saturation. The final doubling adds as many cells as all previous doublings combined.
Fact 2. The growth looks slow until it looks unstoppable. For the first 15 hours of the experiment, the colony is so small that it is invisible to the naked eye. It feels like nothing is happening. Between hours 20 and 30, it jumps from cloudy to saturated. The long, boring prelude is exactly how every exponential process fools the eye.
These are the same two facts that made the chessboard rice story feel like a trick. In bacteria, they are not a parable. A laboratory stopwatch and a spectrophotometer will verify them to three decimal places.
Where this shows up in the real world
Exponential population models show up anywhere that the rate of increase is proportional to the current amount — which turns out to be a lot of places.
- Viral spread. A virus with basic reproduction number R_{0} = 3 triples every generation. After 10 generations, one infected person has seeded 3^{10} \approx 59{,}000 cases. After 20 generations, 3^{20} \approx 3.5 \times 10^{9} — more people than have ever lived in a given country. This is why early containment matters.
- Yeast in bread dough. The warm dough is the nutrient-rich broth. Yeast doubles about every 90 minutes. Over a 3-hour rise, the population quadruples, which is exactly the volume of carbon-dioxide bubbles you need for the dough to puff up.
- Cancer tumour growth (in its early phase). Many tumours double in volume every 60–200 days in the exponential phase, before blood supply runs out and growth slows — the same "logistic" pattern as bacteria in a dish.
- The oxygenation of Earth's atmosphere, billions of years ago. Early photosynthesising cyanobacteria filled the oceans with oxygen in roughly the same pattern: invisible for a long time, and then a sudden atmospheric-scale jump. That jump is the Great Oxygenation Event, and it reshaped Earth's chemistry.
In all of these cases, the exponent is the clock and the base is the per-generation multiplier. Once you recognise the shape, you stop being surprised by the speed.
The takeaway
- A colony that doubles every hour reaches a billion cells in roughly 30 hours, not days or weeks.
- On a \log_{2} axis, exponential growth looks like a straight line — the slope is the doubling rate.
- The last doubling before saturation adds as much as all previous doublings put together. If you can see the dish is half-full, you have one more generation of room.
- Real populations eventually meet a carrying capacity and flatten, but the exponential phase is what every physical intuition should be tuned to recognise.
The petri dish is the laboratory version of the chessboard. It is the experimental confirmation that nature actually does do this, and the reason every biologist, epidemiologist, and cancer researcher reaches for 2^{t} — and the laws of exponents behind it — the moment a population starts to grow.
Related: Exponents and Powers · Grain of Rice on a Chessboard: 64 Squares, Doubling Every Step, Pile Dwarfs Mount Everest · A Tower of 2s: Watch Doubling Explode Off the Screen by Step 30 · Exponent Slider: Watch 2^x Sweep Through 1/8, 1/4, 1/2, 1, 2, 4, 8