This paper presents a method for the comparative analysis of genomes and metabolic pathways based on similarity between gene orders and enzymatic reactions. To measure the reaction similarity, we formalized a scoring system by using the functional hierarchy of the EC numbers of enzymes. We have used an alignment method between given pathways, which is based on the longest common subsequence algorithm with the scoring system. The similarity score between pathways is expressed as the information content of their alignment. By applying our algorithm to the metabolic pathway in Escherichia coli, we have found several common patterns among the purine, lysine and arginine biosynthesis and other amino acid related metabolic pathways. We have also compared the alignments with gene orders on the E. coli genome by using a heuristic graph comparison method. From the comparison, we have found that reaction orders are conserved in the histidine and tryptophan biosynthesis pathways.