Ninety-one Ontario Holstein dairy herds were surveyed about their lactating cow ration and use of a premix containing monensin to identify possible dietary interactions with monensin on milk fat suppression. All herds were enrolled in Ontario Dairy Herd Improvement (DHI) milk recording, and results from four DHI tests were used. Herd mean fat tests were calculated only for cows between 100 and 200 d in milk to avoid potential confounding due to stage of lactation. Wet forage and total mixed ration (TMR) samples from all herds were evaluated for particle size using the Penn State Particle Size Separator. Of the herds using monensin (n = 58), the dose (per kg of dry matter) ranged from 9 to 14 mg/kg in TMR-fed herds and from approximately 9 to 23 mg/kg in herds in which concentrates were fed separately from forages (component-fed). Of the samples submitted for particle size evaluation, 15% of the haylage (n = 80), 14% of the corn silage (n = 79), and 42% of the TMR (n = 58) samples were classified as having low fiber. There was a significant negative univariable association between monensin and mean milk fat percentage. Monensin significantly reduced milk fat percentage in TMR-fed but not component-fed herds. Fiber length significantly interacted with monensin in TMR-fed herds: Herds that had low fiber in their TMR (< or = 6.0% in the top screen) were susceptible to milk fat decrease by monensin, whereas herds that had adequate fiber (> 6.0%) were not. Monensin also significantly reduced milk fat percentage in herds receiving diets low in nonstructural carbohydrate (< 40.2%) but not in those receiving diets high in NSC (> or = 40.2%). The results of this study suggest that there are significant interactions between monensin and certain dietary factors on milk fat suppression in Holstein dairy herds.