The effects of initial flow conditions on the primary breakup of nonturbulent and turbulent round liquid jets in still gases was studied experimentally. Pressure-atomized jets were provided by a piston/cylinder arrangement followed by a converging passage to yield a nonturbulent flow. The degree of flow development at the jet exit was controlled by removing the boundary layer formed along the converging passage, and providing constant diameter passages of various lengths after boundary layer removal. Test conditions included water, n-heptane, and various glycerol mixtures injected into helium, air, and Freon 12 at pressures of 1 and 2 atm, to yield liquid/gas density ratios in the range 104-7240. Pulsed photography and holography were used to observe the liquid surface prior to primary breakup. The results highlight the importance of liquid vorticity at the jet exit on primary breakup at these conditions: Experiments with nearly vorticity-free exit conditions (passage length/diameter ratio, L/d = 0.15) caused primary breakup to be suppressed, yielding stable liquid jets similar to those used in liquid jet cutting processes. In contrast, larger L/d at sufficiently high Reynolds numbers caused transition to turbulent jets having wrinkled surfaces prior to primary breakup by the turbulent primary breakup mechanism. A breakup regime map was developed, yielding behavior in the turbulent primary breakup regime for L/d > 4-6 and passage Reynolds numbers > 1-4 x 10(4). Within the turbulent primary Breakup regime, conditions for the onset of breakup and the evolution of drop sizes with distance from the jet exit were relatively independent of L/d for values up to 212. Finally, a new correlation for drop sizes after primary breakup at nonturbulent conditions (but with laminar boundary layers present along the passage wall) was developed, based on consideration of boundary-layer thicknesses at the jet exit.