The effects of supplementation with grass silage and replacement of some corn in the concentrate with soybean meal (SBM) on milk production, and milk fatty acid (FA) profiles were evaluated in a replicated 4 x 4 Latin square study using 16 dairy cows grazing pasture composed of ryegrass, Kentucky bluegrass, and white clover. Each experimental period lasted for 3 wk. The 4 dietary treatments were PC, 20 h of access to grazing pasture, supplemented with 6 kg/d of corn-based concentrate mixture (96% corn; C); PCSB, 20 h of access to grazing pasture, supplemented with 6 kg/d of corn- and SBM-based concentrate mixture (78% corn and 18% SBM; CSB); SC, 7 h of access to grazing pasture during the day and 13 h of ad libitum access to grass silage at night, supplemented with 6 kg/d of C concentrate; and SCSB, 7 h of access to grazing pasture during the day and 13 h of ad libitum access to grass silage at night, supplemented with 6 kg/d of CSB concentrate. The concentrate mixtures were offered twice each day in the milking parlor and were consumed completely. Grass silage supplementation reduced dietary crude protein and concentration of total sugars, and dietary SBM inclusion increased dietary crude protein concentration and decreased dietary starch concentration. Milk yield and energy-corrected milk were increased by SBM supplementation of cows with access to grass silage. Milk protein concentration was lower in cows offered grass silage, regardless of whether SBM was fed. Dietary SBM inclusion tended to increase milk fat concentration. Plasma urea N was reduced by silage feeding and increased by SBM supplementation. Supplementation with grass silage overnight could represent a useful strategy for periods of lower pasture availability. Dietary inclusion of SBM in solely grazing cows had no effects on milk production and composition, exacerbated the inefficient capture of dietary N, and increased diet cost. Grass silage supplementation affected milk FA profiles, increasing both the FA derived from de novo synthesis and those derived from rumen microbial biomass, and decreasing the sum of C18 FA (mostly derived from diet or from mobilization of adipose tissue reserves). Milk fat concentrations of conjugated linoleic acid cis-9, trans-11, vaccenic acid (18:1 trans-11), and linolenic acid (18:3n-3) were unaffected by grass silage supplementation, suggesting that partial replacement of pasture by unwilted grass silage does not compromise the dietary quality of milk fat for humans.