Evolution of novel metabolic pathways drives organismal diversity by enabling access to alternative nitrogen or carbon sources or facilitating synthesis of beneficial compounds. Bioinformatic evidence suggests that new metabolic pathways are assembled by recruiting promiscuous enzymes to serve new functions. Enzyme promiscuity provides evolutionary footholds when new reactions become important for fitness. How flux is increased through promiscuous activities to establish a novel metabolic pathway is unknown. The Copley lab previously characterized a novel metabolic pathway that evolved in an engineered strain of Escherichia coli to restore synthesis of the essential metabolite pyridoxal 5′-phosphate (PLP). One evolved strain, Ec1, contained four mutations. To map the fitness landscape for evolving a new metabolic pathway, I characterized a set of sixteen strains comprised of the unevolved strain with a disrupted PLP synthesis pathway, Ec1 and all fourteen intermediates. I found that a mutation shifting glycolytic flux was critical for increasing flux through the new pathway. Comparison of this fitness landscape to the actual trajectory by which Ec1 evolved showed that it evolved via an intermediate “cheater” strain that was less fit than its progenitor. This case study provides one example of a novel metabolic pathway evolving. However, differences between bacteria could impact how metabolic pathways evolve. The repertoire of promiscuous enzymes and their activities varies across species. These differences should affect the ability of promiscuous enzymes iii to be recruited to a new pathway. To investigate how these differences impact metabolic pathway evolution, I expanded the model system developed in E. coli to Allivibrio fischeri, Pseudomonas putida and Salmonella enterica. In collaboration with other members of the Copley lab, I showed that some of these organisms evolve different pathways than E. coli and that even when they evolve the same pathways, they do so by different mutations. Together, these studies illustrate how flux can be increased through a novel metabolic pathway and how genomic and proteomic differences impact this process.