Genetically identical cells that live in a homogeneous environment often show substantial variation in their biological traits; such variation is called phenotypic noise. The level of phenotypic noise has a genetic basis, suggesting that higher levels of phenotypic noise can evolve. In fact, recent theoretical studies suggest that phenotypic noise could be a mechanism for a population of genotypes to respond to uncertain environments in a more efficient way than with conventional signal transduction pathways. However, it is not known if phenotypic noise is relevant for bacterially-driven processes in the environment, because the few studies that experimentally investigated phenotypic noise in bacteria did not consider metabolic activities that contribute to biogeochemical cycles. While it has been observed with novel nanoSIMS (nano-scale secondary ion mass spectrometry) technology that bacteria display phenotypic noise in metabolic activities, direct experimental evidence that phenotypic noise in metabolic activities has a biological function and can provide isogenic bacterial populations with a growth advantage is missing on a single-cell level. Moreover, it has not been experimentally tested if phenotypic noise in metabolic activities adapts over evolutionary timescales in response to fluctuating environmental conditions. The goal of this project is to experimentally investigate how phenotypic noise affects bacterial metabolic activity, growth and evolution under fluctuating environmental conditions. The proposal focuses on phenotypic noise in N2-fixation in the unicellular aquatic bacterium Klebsiella pneumoniae. The experiments will combine time-lapse microscopy, nanoSIMS and experimental evolution to understand why bacteria display phenotypic noise in metabolic activities. This will establish a link between the behaviour of single cells and biogeochemical cycles, and reveal how variation at the single-cell level can impact processes at the ecosystem level.