Skip to content

Two microbes, one dish

A teaching toy, not a research result. This page has nothing to do with irit’s command-line tools. It shows a textbook model of two species competing for the same resources, so you can see how growth rates and competition strengths decide whether one species wins or both survive.

The model is the two-species competitive Lotka–Volterra system, solved in your browser with a fourth-order Runge–Kutta integrator:

dN₁/dt = r₁N₁ (1 − (N₁ + a₁₂N₂) / K₁)
dN₂/dt = r₂N₂ (1 − (N₂ + a₂₁N₁) / K₂)

N is a population size, r its growth rate when rare, K the size it settles at when alone (the carrying capacity), and a₁₂ how much one individual of species 2 counts against species 1’s capacity (a₂₁ the reverse). The two preset experiments used yeasts and the ciliate Paramecium: those are the classic data sets where every parameter was published. The rods and rounds in the dish are cartoons, not drawings of those organisms.

Time shown

The dish

Cartoon dish. The number of rod-shaped and round characters follows the two populations at the time shown below.

Populations over time
Time axis
Phase plane

Each species' zero-growth line (nullcline) and the path the two populations take. Click or tap to start from another point.

Parameters

Moving a slider switches to Explore mode.

Four possible endings

  1. Species 1 winsCurrent

    K₁ > a₁₂K₂ and K₂ < a₂₁K₁

    Species 1 can grow when it is rare and species 2 cannot. Species 2 dies out from any start.

  2. Species 2 winsCurrent

    K₁ < a₁₂K₂ and K₂ > a₂₁K₁

    The mirror image: species 2 can invade, species 1 cannot.

  3. Both coexistCurrent

    K₁ > a₁₂K₂ and K₂ > a₂₁K₁

    Each species holds itself back more than it holds back the other, so either can recover when rare. Both settle at a shared balance point.

  4. Whoever leads winsCurrent

    K₁ < a₁₂K₂ and K₂ < a₂₁K₁

    Each species holds the other back more than itself. Neither can invade the other, so the winner depends on where they start.

  • The dish shows the populations at the time on the slider: more characters means a larger population, scaled so the larger carrying capacity fills the dish. Press Play to watch the run from the start.
  • Populations over time: solid line for species 1, dashed for species 2, with each one’s carrying capacity as a dotted line.
  • Phase plane: each nullcline is where one species stops growing. Filled circles are stable end points, open circles unstable ones. Where the two lines sit decides which of the four endings you get.

Real cultures have lag phases, toxins, spatial structure, mutation and chance. Biochemical factors (nutrients, metabolites, pH) are held constant here; the model folds them into r, K and a. Those parameters are constants too, while in Gause’s own data they changed as the cultures grew. Treat the presets as a way to see the model behave on real numbers, not as predictions about these organisms.