The objective was to determine whether the use of estradiol-17beta (E-2) at the initiation of short dry periods prevented an anticipated decline in milk production in the subsequent lactation. Lactating Holstein cows (n = 66) were dried at either 60 or 30 d before expected calving. Treatments in a 2 x 2 factorial arrangement included: D60 (n = 19, 60-d dry, no E-2), D60 + E-2 (n = 18, 60-d dry, E-2), D30 (n = 15, 30-d dry, no E-2), and D30 + E-2 (n = 14, 30-d dry, E-2). To accelerate mammary involution, estradiol-17beta (15 mg in 4 ml of ethanol) was injected subcutaneously daily for 4 d beginning 30 d before expected calving. Parturitions occurred between November 1995, and March 1996. Actual days dry for respective treatments were 57.3, 60.6, 33.9, and 33.8 +/- 1.7 d. Onset of parturition, calving difficulty, and cow health were not affected by E-2. Actual 305-d milk yields for the lactation completed immediately before the experimental dry period were 10,318, 10,635, 10,127, and 10,447 +/- 334 kg, respectively; and were 9942, 9887, 9669, and 10,172 +/- 387 kg, respectively, for the lactation immediately following treatment. Respective pre- and posttreatment mature equivalent 305-d yields were 9574, 9861, 9812, and 9724 +/- 297 kg; 8987, 8843, 9126, and 9008 +/- 294 kg. Milk yields did not differ across treatments. Cows with a 34-d dry period were as productive as cows with a 59-d dry period. Estradiol-17beta had no effect, but perhaps should be evaluated with dry periods shorter than 34 d.