The response of dairy cows to monensin can interact with diet composition. We evaluated the response of lactating cows to monensin supplementation of diets differing in rumen fermentable starch. Twenty-eight Holsteins (157 ± 76 DIM) were individually fed a standard TMR for 3 wks and monensin (MON. 300 mg/d) or control (CTL) for 9 wks, in a covariate adjusted randomized block design. A low-starch diet (LS. 25.8% starch in DM) was offered from d 1 to 35 and a high-starch diet (HS. 30% starch in DM) from d 36 to 63. HS was formulated by replacing whole cottonseed and finely ground mature corn by an isonitrogenous mixture of soybean meal and high moisture corn. Data were collected daily or after adaptation to LS and HS, and were analyzed as repeated measures with PROC MIXED of SAS. The mean square for cow nested within treatment (MON vs. CTL) was the error term for the treatment effect. Digestible OM intake (DOMI) was increased by HS (15.3 vs. 14.4 kg/d. P < 0.01), but was similar for MON (14.6 kg/d) and CTL (15.1 kg/d. P = 0.41). Milk yield did not respond to MON (32.3 kg/d. P = 0.32). MON reduced DMI and increased Milk/DMI, ECM/DMI, and ECM/DOMI only when HS was fed (P < 0.01 for the interaction of treatment and wk/diet). Plasma glucose content was increased by MON and the response interacted similarly with wk/diet (P < 0.05). MON reduced the A/P ratio in rumen fluid in both diets (2.43 vs. 1.81. P < 0.01) and HS reduced ruminal pH (P < 0.01) and increased protozoa content (P < 0.01). MON reduced total-tract NDF digestibility more in LS (52.1 vs. 41.3% of intake) than in HS (57.0 vs. 52.0. P < 0.05 for the interaction). The daily urinary allantoin excretion did not respond to MON (P = 0.40), neither PUN at 0, 1.5, and 3 h post feeding (P > 0.27). MON induced lower degree of refusal of long feed particles and of preferential intake of short particles in the afternoon than CTL (P < 0.05). Ingestion time (min/d and min/kg of DMI) and the number of daily meals were reduced by HS (P < 0.05). HS increased pCO₂, HCO₃^-, total CO₂, and base excess of jugular blood (P < 0.04) and tended to reduce blood pH (P = 0.10). The positive effect of MON on feed efficiency occurred only when ruminally available starch was increased in the diet.