Eighty-nine samples, 45 of standing forage and 44 of baled hay, were collected from alfalfa harvested at various maturities over three cuttings each during 2 yr. Alfalfa was cut and conditioned mechanically; samples of standing forage were collected by removing bunches of forage from windrows and freeze-drying them. Forage was allowed to field cure and was harvested at an average 80% DM as small rectangular bales; samples of baled hay were collected by coring bales after storing for 3 to 6 mo. Samples were analyzed for DM, ADF, total N, fractions of total N present as ADIN, N degraded at 0 h, and potentially degradable protein N. Ruminal protein degradation rates and escapes were estimated using an inhibitor in vitro system, assuming that ADIN was unavailable and that ruminal passage rate was .06/h. Standing forage contained smaller fractions of ADIN and N degraded at 0 h, contained a larger fraction of potentially degradable N, and had more rapid degradation rates and lower estimated protein escapes than baled hay. Mean degradation rates and estimated escapes were .171/h and 24% for standing forage and .075/h and 40% for baled hay. There were no differences in degradation rate or estimated escape because of harvest year, and neither was significantly related to maturity or to ADF concentration. Results indicate a significant advantage in ruminal protein escape, compared with grazed alfalfa, for alfalfa harvested and stored as hay.