In a simple model (top), microbes (N) grow exponentially at rate μ, while each immune cell or effector molecule (I) can kill a microbe at rate γ. The number of immune cells or molecules depends on the rate λ, which in turn depends on a number of factors such as the level of standing immunity, the time taken to activate the immune response (bottom left), and the maximum capacity of the system to produce immune components. The interaction between a given type of microbe and the immune system can be characterised by a map (bottom right) that plots μN – γI as a function of I (horizontal axis) and N (vertical axis). If the growth rate of microbes exceeds the killing capacity of the immune system (μN > γI), the microbes grow exponentially throughout the infection (solid line; red region). If the growth rates become equal (μN = γI), the outcome is a persistent, 'set-point' level of chronic infection (long dashes; white region). If the growth rate of the immune response outpaces that of microbes (μN < γI), the infection will clear up (short dashes; blue) – this was, however, rarely observed in the fruit flies. Duneau et al. suggest that if the number of microbes reaches a certain level (Ntip), the host will become incapable of producing a sufficient immune response to control the microbes. In this case, the bacteria will continue to grow until the host is dead.