The objective of this study was to determine the effects of altering dry period length in multiparous dairy cows (n=341) on milk production for a full lactation (294 d). The study used 3 commercial herds in the western United States. Cows producing greater than 8,400kg of mature-equivalent milk were assigned to treatments 60 d before their due dates. The 4 treatments were 1) 60-d dry period, label use of recombinant bovine somatotropin (rbST; 60d-L); 2) 32-d dry period, label use of rbST (32d-L); 3) 0-d dry period, label use of rbST (0d-L); and 4) 0-d dry period, continuous use of rbST (0d-C). Cows with shortened dry periods produced 625, 1,000, and 1,042kg of milk during the prepartum period for treatments 2 to 4, respectively. During the postpartum period, cows on the 32d-L treatment produced similar amounts of milk compared with the 60d-L treatment (11,490 vs. 11,968kg, respectively). However, cows on the 0d-L (10,316kg) and 0d-C (10,195kg) treatments produced significantly lower amounts of milk during the postpartum period compared with the 60d-L treatment. Total milk production from the prepartum and postpartum periods was not altered significantly and was 11,974, 12,112, 11,310, and 11,230kg for treatments 1 to 4, respectively. The concentrations of β-hydroxybutyrate and nonesterified fatty acids in serum after calving were decreased for cows on the 32d-L, 0d-L, and 0d-C treatments compared with cows on the 60d-L treatment, which may indicate improved metabolic status.