How does a trait evolve from A to B? Does it take many small steps, or one big one, or does it take one largish step followed by a few small ones? These questions are difficult to answer, mainly because adaptive events are only observed after they have taken place. An experimental evolution study in Pseudomonas spp. has captured the first adaptive event as it happens — the fitness advantage of such mutations is high, and so that first step to adaptation is more of a jump.

The view that evolution is a gradual process has been challenged by evidence that large-effect mutations can also underlie adaptive changes. But developing a general rule of adaptation is not easy, especially because beneficial mutations occur rarely. Experimental evolution presents a unique advantage: many beneficial mutations can be recovered, and their adaptive fitness can be compared to that of the ancestral population.

In this study, populations of the bacterium Pseudomonas fluorescens were grown in a harsh, carbon-limited medium for approximately 100 generations. The selection coefficients of the 68 fixed beneficial mutations that were recovered followed a bell-curve distribution; they also occurred at a higher rate than expected (3.8 × 10−8 per cell division) and had a much greater selective advantage (2.1).

The fact that the first beneficial steps in evolution are larger than previously supposed has theoretical importance, but it also has a direct bearing on modelling the evolutionary trajectory of microbes that are harmful to human health.