A plant needs both phosphorus and nitrogen. Both are in the soil. But phosphorus barely moves — it binds to soil particles and diffuses at rates measured in millimeters per year. Nitrogen, as nitrate, dissolves in water and percolates downward with every rain. Same soil, same plant, same roots. But the geometry that captures one resource is wrong for the other.
Jonathan Lynch and colleagues described the solution. For phosphorus: shallow root angles, dense lateral branching, proliferating root hairs. The strategy is topsoil foraging — maximizing surface area in the upper centimeters where organic matter decomposes and releases phosphorus into a thin, nearly immobile layer. Root hairs alone can increase absorptive surface area tenfold. The architecture is fractal: branch, rebranch, subdivide.
For nitrogen and water: steep angles, minimal branching, thin axial roots. Lynch called it “steep, cheap, and deep.” Nitrate leaches below the topsoil with every rainfall event. The architecture that captures it is penetrative — depth, not surface area. Lateral branches cost metabolic carbon without reaching the resource. The optimal root is a spear, not a net.
The structural insight is that foraging geometry is determined by resource mobility, not resource location. Both nutrients are “in the soil.” But immobile resources demand fractal expansion — maximum contact area in a confined zone. Mobile resources demand linear extension — maximum reach along the axis of flow. The root system is not searching randomly; it is predicting how its target moves, and building the geometry that intercepts that movement.
This generalizes. Any system that searches for something must match its search geometry to the target's dynamics. Casting a wide net works for stationary targets. Drilling deep works for flowing ones. The architecture of pursuit is a physical encoding of the prey's behavior — a prediction made structural.