Why food rots

From salting and drying to pickling and irradiating, humans have devised many ingenious ways of preserving their food from spoilage by microbes. The question of what microbes gain from making food go off in the first place has attracted less attention, but research presented at this year's British Ecological Society Annual Meeting shed new light on the problem.

Speaking at the meeting, Dr Dave Wilkinson of Liverpool John Moores University and Dr Thomas Sherratt of Carleton University in Canada cast doubt over Professor Dan Janzen's seductive 1977 theory that microbes make food go off in order to make it objectionable or unusable by the larger animals they are competing with for food.

Janzen illustrated his theory thus: imagine a child left alone for a short time in the kitchen with strawberries, one fresh and one mouldy. If the youngster pops the fresh one into its mouth, then the microbe has won.

Wilkinson and Sherratt used mathematical models for the first time to test Janzen's theory. According to Wilkinson: "In our current model it is difficult to see how spoiling behaviour could evolve as an adaptation to deter larger animals. Janzen's idea, while intuitively attractive, may be unworkable. Our main result is that in the mathematical model we have developed so far, we have been unable to find realistic conditions under which cheats will not undermine the system.

"If microbes are expending energy producing chemicals to deter birds and other animals from eating their food, then what is to stop them from cheating by not producing the chemical but just relying on protection from chemicals produced by other microbes?"

As well as challenging Janzen's theory, Wilkinson and Sherratt's work could help ecologists understand cheating in other areas of nature. "The problem of cheats destroying systems of mutually cooperating organisms is a major problem in evolutionary ecology. Janzen's system has great merit because it is relatively simple and therefore open to mathematical study, so it may yet help identify the conditions under which cooperation is favoured over cheating," Wilkinson says.

Wilkinson and Sherratt are currently building more complex models to see if this helps rescue Janzen's theory.