Bacillus species produce extracellular, surface-active lipopeptides such as surfactin that have wide applications in industry and medicine. The steps involved in the synthesis of 3-hydroxyacyl-coenzyme A (CoA) substrates needed for surfactin biosynthesis are not understood. Cell-free extracts of Bacillus subtilis strain OKB105 synthesized lipopeptide biosurfactants in presence of l-amino acids, myristic acid, coenzyme A, ATP, and H2O2, which suggested that 3-hydroxylation occurs prior to CoA ligation of the long chain fatty acids (LCFAs). We hypothesized that YbdT, a cytochrome P450 enzyme known to beta-hydroxylate LCFAs, functions to form 3-hydroxy fatty acids for lipopeptide biosynthesis. An in-frame mutation of ybdT was constructed and the resulting mutant strain (NHY1) produced predominantly non-hydroxylated lipopeptide with diminished biosurfactant and beta-hemolytic activities. Mass spectrometry showed that 95.6% of the fatty acids in the NHY1 biosurfactant were non-hydroxylated compared to only ∼61% in the OKB105 biosurfactant. Cell-free extracts of the NHY1 synthesized surfactin containing 3-hydroxymyristic acid from 3-hydroxymyristoyl-CoA at a specific activity similar to that of the wild type (17 ± 2 versus 17.4 ± 6 ng biosurfactant min−1·ng·protein−1, respectively). These results showed that the mutation did not affect any function needed to synthesize surfactin once the 3-hydroxyacyl-CoA substrate was formed and that YbdT functions to supply 3-hydroxy fatty acid for surfactin biosynthesis. The fact that YbdT is a peroxidase could explain why biosurfactant production is rarely observed in anaerobically grown Bacillus species. Manipulation of LCFA specificity of YbdT could provide a new route to produce biosurfactants with activities tailored to specific functions.