Bubble formation is monitored in a reactor such as a loop polymerization reactor which effectively operates at pressures above or near the vapor pressure of a circulating slurry liquid. The method measures a property of the reactor composition at a first location within the reactor, and then at a second location within the reactor. The process is repeated, followed by determining if the difference in measured property varies from a reference difference, derived under bubble-free conditions. When the difference in measured physical property has changed relative to the reference difference, an effect may be implemented, e.g., increase of reactor pressure, decrease of composition vapor pressure, and decrease of reactor temperature, to an extent sufficient to reduce or eliminate bubbling.