Two experiments were conducted to ascertain lysine required for maintenance and milk production in lactating dairy cows. Multiparous Holstein cows fitted with abomasal cannulas were utilized in a replicated 3 x 3 extra-period Latin square design and infused with 0, 45, or 90 g of L-lysine-HCl daily (Experiment 1). In Experiment 2, 12 cows were utilized in a 5 x 5 Latin square design (with 2 replacement cows) and infused with 0, 22.5, 45, 90, and 180 g of L-lysine-HCl daily. Cows were housed in comfort stalls, milked at 0600 and 1700 h, exposed to a 16:8 h light:darkness cycle, and fed for ad libitum intake a 15.7% CP diet formulated to contain approximately 75% of dietary protein from corn-based feed ingredients. Milk yield was not affected in Experiment 1, but milk production increased linearly with lysine infusion level in Experiment 2. A linear response in daily yield of milk protein to lysine infusion was observed in both experiments. As lysine infusion increased, coccygeal plasma lysine concentrations tended to increase in Experiment 1, but in Experiment 2 free lysine increased linearly and quadratically in coccygeal plasma and increased linearly in plasma from the subcutaneous abdominal vein. The quadratic response in plasma free lysine (Experiment 2) suggested that lysine was the limiting AA for milk production from the composite protein flow to the small intestine. Using break-point analysis of plasma free lysine, it was estimated that the additional amount of abomasal lysine required to satisfy requirements was 64 g/d in cows producing about 30 kg of milk daily. Although additional lysine above microbial and undegraded intake protein lysine flow stimulated milk protein production, conversion efficiency to milk protein lysine was low.