Glyphosate and adjuvant combinations were applied to rhizome johnsongrass at vegetative and reproductive growth stages to evaluate control and rainfastness in field studies. Using a rainfall simulator delivering 1.3 cm of water in 15 min, plots received either no rainfall or rainfall 15 or 60 min after glyphosate was applied at 2.1 kg ai/ha in combination with the nonionic surfactants Kinetic® HV at 0.25% (v/v) or Induce® at 1.0% (v/v) or the silicone surfactant Break-Thru® at 0.125% (v/v). Regardless of adjuvant, rainfall 15 or 60 min after application reduced johnsongrass control compared with no rainfall. Johnsongrass control 14 d after treatment at the reproductive stage was at least 89% with no rainfall, but no more than 53 and 65% with rainfall at 15 and 60 min, respectively. Based on initial weed control, adjuvants did not consistently improve rainfastness. Johnsongrass regrowth did not occur when glyphosate was applied with either adjuvant. In contrast, for glyphosate applied to johnsongrass in the vegetative stage, addition of Break-Thru improved control over Induce at both 15- and 60-min rainfall timings in one of two experiments. With no rainfall, addition of Kinetic HV and Break-Thru increased johnsongrass control in only one experiment. For application at the vegetative stage, johnsongrass regrowth averaged across rainfall timings was no more than 10%. In other field experiments, glyphosate at 1.4 kg/ha plus nonionic surfactants, silicone surfactant, crop oil concentrate, methylated seed oil, or a blend of silicone surfactant and methylated seed oil were equally effective in reducing johnsongrass regrowth when applied after seedhead emergence. Improved control of vegetative johnsongrass with some adjuvants was not reflected in decreased regrowth.