Field trials were initiated during 2004 at a dryland site near Prosper, N.D., to evaluate the effects of simulated drift from glyphosate to `Russet Burbank' and `Red Lasoda' seed potato during the early senescence stage. Glyphosate was applied at rates 1/3, 1/6, 1/12, 1/24, and 1/48 the use rate for spring wheat desiccation on 10 Sept. 2004 with a CO2 pressurized sprayer operated at 280 L/ha and 276 kPa. The amount of a.m.S added to the spray solution was also reduced accordingly. Following harvest, samples from each plot were placed into cold storage until the following March. A subsample from each plot was slowly warmed to initiate sprout formation. Remaining samples were cut into 57-g pieces, dusted with a seed piece treatment, and stored at 18 °C with about 90% RH until planted. Plots consisted of two 3-m rows at 91 cm-row spacing with a border row on each side and three spacer plants between plots. The trial was arranged as a randomized complete block with four replications. Plots were desiccated on 12 and 19 Sept. and harvested 11 Oct. Tubers were hand-graded shortly after harvest. Results indicated that glyphosate at 70 g ae/ha or more inhibited tuber budbreak by 75% or more compared to untreated. In the field, injury was observed as delayed emergence and, in several instances, no plants emerged. Total yield for `Red Lasoda' was 34.8 Mg/ha for the untreated, which was significantly greater than glyphosate treatments of 280, 140, and 70 g ae/ha. `Russet Burbank' total yield was considerably less at 23.5 Mg/ha for the untreated. Both the untreated and glyphosate at 18 g ae/ha had significantly greater total yields compared with glyphosate treatments of 280, 140, and 70 g ae/ha.