Laboratory bacterial populations provide ideal opportunities to experimentally test theories in ecology and evolutionary biology. I used a model laboratory microbial system, Pseudomonas fluorescens SBW25, to address an array of questions on the origin, maintenance, and functional role of biodiversity, and the evolution of biotic interactions. My thesis reports experiments with the following conclusions. (1) The extent of diversification in P. fluorescens populations is not affected by the presence of an interspecific competitor P. putida, although the early stage of the diversification in one environment (spatially homogeneous environment) could be speeded up by the competitor. (2) Niche and neutral mechanisms simultaneously contribute to the maintenance of phenotypic diversity in P. fluorescens populations; but the operation of niche processes does not lead to a positive effect of biodiversity on ecosystem functioning. (3) The competitive interactions among bacterial phenotypes are generally transitive, and competitive hierarchies inferred from pair-wise competition are fairly consistent to those from multi-species competition. (4) The niche complementarity and selection effects evaluated by random assembly biodiversity experiments can be used to predict the functional consequences of particular non-random species extinction scenarios. (5) P. fluorescens does not show an evolutionary trade-off in using several carbon substrates (glucose, galactose and trehalose), and evolution in environments containing these resources results in imperfect generalists; migration among populations may speed up fitness evolution of some generalists. (6) Biofilm formation at the air-broth interface by wrinkly spreader phenotypes in P. fluorescens is a cooperative behaviour which is costly to individuals but benefits the group; this behaviour could be exploited by smooth morph phenotypes. The cooperators and cheats in this system show reciprocal antagonistic coevolution in resistance and cheating performance.