My interest in the evolution of diversity in microbial populations began more than 20 years ago. For the first 10 of those years I was oblivious to the fact that one of the most dramatic forms to emerge during the course of selection experiments, the so-named wrinkly spreader (WS) type, owed its success to cooperation among individual cells. Rather ashamedly, despite having recognised the novelty of what I had witnessed, it took me another 10 years to get round to publishing this work. Perhaps, however, an attempt to publish in the early 1990s, in the absence of studies that gave credibility to the microcosm experiments (see Chapter 13), would have met with limited success.

There was no eureka moment of realisation, although with hindsight there ought to have been. I was aware that WS genotypes formed cellular mats that grew at the air–liquid interface of broth-filled microcosms (Rainey & Travisano 1998). I was also aware that the ability to occupy the air–liquid interface was the secret of their evolutionary success (the broth phase rapidly become anaerobic due to microbial growth). Most tellingly, I was aware that the mats sank into the broth when they became old and heavy.