The graph that best represents disruptive selection shows a bell curve that splits into two peaks at opposite ends of the trait range, with a valley (dip) in the middle. If you’re looking at a multiple-choice question, pick the graph where the original single-humped curve becomes a two-humped, bimodal curve, with fewer individuals showing intermediate trait values and more individuals at both extremes.
What the Graph Looks Like
A standard disruptive selection graph has “Trait Value” on the x-axis and “Frequency” (number of individuals) on the y-axis. The starting population is typically drawn as a normal bell curve. After disruptive selection acts on the population, the curve transforms: the middle section drops down while both tails rise, creating two distinct peaks. This bimodal shape is the signature of disruptive selection and the single most important visual feature to look for.
Think of it like a camel’s back with two humps. The left hump represents individuals with small or low trait values, the right hump represents individuals with large or high trait values, and the dip between them shows that average individuals are being selected against. The overall variance of the population increases because the trait values are now spread across a wider range than before.
How It Differs From the Other Two Types
On a typical exam or textbook figure, you’ll see three selection types side by side. Here’s how to tell them apart at a glance:
- Stabilizing selection: The bell curve gets narrower and taller. Both extremes lose individuals, and the average trait value stays the same. The peak sharpens in the center.
- Directional selection: The entire bell curve shifts left or right, toward one extreme. The peak moves, and one tail shrinks while the other grows. The curve may also appear skewed.
- Disruptive selection: The bell curve splits into two peaks. The center dips. Both extremes gain individuals at the expense of the middle.
If you remember just one rule, it’s this: stabilizing selection removes the extremes, directional selection removes one extreme, and disruptive selection removes the middle.
Why the Middle Drops Out
Disruptive selection happens when individuals at both ends of a trait spectrum have higher fitness than those in between. The environment essentially has two “winning strategies,” and being average is the worst option. A classic example comes from Darwin’s finches, where disruptive selection was observed acting on beak size. Birds with very large beaks could crack hard seeds, and birds with very small beaks could efficiently handle tiny seeds, but medium-beaked birds were outcompeted in both niches. Selection was strong between the two beak size modes, maintaining the population’s bimodal distribution.
In fitness landscape terms, the environment creates a concave (valley-shaped) surface where the population mean sits at a low point between two fitness peaks. Individuals closer to either peak survive and reproduce at higher rates, pulling the population apart over generations.
What Changes Statistically
The key statistical change under disruptive selection is increased variance. The mean trait value of the population may not shift much, since both extremes are favored roughly equally. But the spread of trait values gets wider. Research on pike in England’s Lake Windermere demonstrated this clearly: when fishing practices created disruptive selection on body size (harvesting medium-sized fish while leaving the smallest and largest), both slow-growing and fast-growing fish were simultaneously favored. Phenotypic variance in growth rate increased over nearly two decades, exactly as theory predicts.
This is the opposite of stabilizing selection, which narrows variance, and different from directional selection, which shifts the mean.
Why Disruptive Selection Matters for Speciation
Disruptive selection is the only type of natural selection that can split a single population into two. When individuals at opposite extremes of a trait are consistently favored, and if those individuals also start mating preferentially with others like themselves, the population can eventually divide into two separate species without any geographic barrier. This process is called sympatric speciation.
The catch is that random mating normally prevents the split. If large-beaked and small-beaked finches keep mating with each other, their offspring end up with intermediate beaks and low fitness, blending the population back together. Speciation proceeds only when some form of assortative mating evolves, such as a preference for habitats associated with a particular trait. Modeling research has shown that learned habitat preferences can reinforce disruptive selection and make speciation far more likely, because offspring that learn to prefer the habitat they grew up in will tend to encounter and mate with similar individuals.
On a graph, this long-term outcome would appear as the two peaks moving further apart and the valley between them deepening until the two humps represent effectively separate populations.