If Bacillus subtilis is incubated in radioactive methionine in the absence of protein synthesis, the methyl-accepting chemotaxis proteins (MCPs) become radioactively methylated. If the bacteria are further incubated in excess nonradioactive methionine (“cold-chased”) and then given the attractant aspartate, the MCPs lose about half of their radioactivity due to turnover, in which lower specific activity methyl groups from S'-adenosylmethionine (AdoMet) replace higher specific activity ones. Due to the cold-chase, the specific activity of the AdoMet pool is reduced at least 2-fold. If, later, the attractant is removed, higher specific activity methyl groups return to the MCPs. Thus, there must exist an unidentified methyl carrier that can “reversibly” receive methyl groups from the MCPs. In a similar experiment, labeled cells were transferred to a flow cell and exposed to addition and removal of attractant and of repellent. All four kinds of stimuli were found to cause methanol production. Bacteria with maximally labeled MCPs were exposed to many cycles of addition and removal of attractant; the maximum amount of radioactive methanol was evolved on the third, not the first, cycle. This result suggests that there is a precursor-product relationship between methyl groups on the MCPs and on the unidentified carrier, which might be the direct source of methanol. However, since no methanol was produced when a methyltransferase mutant, whose MCPs were un-methylated, was exposed to addition and removal of attractant or repellent, the methanol must ultimately derive from methylated MCPs.