The kinetics of cell cycle initiation were measured at pH 2.7 for cells that had been arrested at the "start" step of cell division with the polypeptide pheromone alpha-factor. Cell cycle initiation was induced by the removal of alpha-factor. The rate at which cells completed start was identical to the rate of subsequent bud emergence. After short times of prearrest with alpha-factor (e.g. 5.2 h), the kinetics of bud emergence were biphasic, indicative of two subpopulations of cells that differed by greater than 10-fold in their rates of cell cycle initiation. The subpopulation that exhibited a slow rate of cell cycle initiation is comprised of cells that resided in G1 prior to start at the time of removal of alpha-factor, whereas the subpopulation that initiated the cell cycle rapidly is comprised of cells that had reached and become blocked at start. A critical concentration of cycloheximide was found to reintroduce slow budding cells into a population of 100% fast budding cells, suggesting that the two subpopulations differ with respect to attainment of a critical rate of protein synthesis that is necessary for the performance of start. Cycloheximide and an increase in the time of prearrest with alpha-factor had opposite effects on both the partitioning of cells between the two subpopulations and the net rate of protein synthesis per cell, consistent with this conclusion. Cell cycle initiation by the subpopulation of fast budding cells required protein synthesis even though the critical rate of protein synthesis had been achieved during arrest. It is concluded that alpha-factor inhibits the synthesis of and/or inactivates specific proteins that are required for the performance of start, but alpha-factor does not prevent attainment of the critical rate of protein synthesis.