Co-expression analysis allows experimenters to re-use archived expression microarray data to uncover previously unknown functional relationships between genes. An observation that a group of genes are co-expressed across diverse experimental conditions suggests they may play similar roles in the cell. Several thousand expression microarray experiments performed on samples from Arabidopsis thaliana have entered the public domain and it is now possible to use these data to investigate metabolic networks in plants. This chapter explains how to use a Web-based tool (CressExpress) to investigate co-expression of genes involved in metabolic pathways in Arabidopsis. Using CressExpress together with desktop visualization and analysis tools, one can easily identify clusters of genes that are co-expressed with one or more genes of interest, making it possible to identify new players in metabolic pathways that are regulated at the level of mRNA abundance.