Animal performance can be improved by overseeding legumes into grass swards, but little is known concerning consumption of various legumes over time. This study compared selection of various legumes to grass and weeds, and gathered information helpful in configuring grazing systems. Over two grazing seasons, esophageally fistulated steers (Bos taurus) sampled replicated pasture systems of; (i) orchardgrass (Dactylis glomerata L.) and red clover (Trifolium pratense L.) (OG‐RC), (ii) orchardgrass and birdsfoot trefoil (Lotus corniculatus L.) (OG‐BFT), and (iii) orchardgrass alone (OG) as spring and fall pasture near Columbia, MO. Steers also sampled the same treatments as summer hay regrowth, except that alfalfa [Medicago sativa L.] replaced birdsfoot trefoil in Treatment ii (OGALF). Grass dominated swards and steer diets, but selectivity ratios (SR) indicated no preferential selection for grass by steers. Dietary amounts of birdsfoot trefoil and red clover on pasture, and red clover and alfalfa on hay regrowth were similar, although availability of red clover was generally less than the other legumes. Thus, steers selected red clover to a greater extent over birdsfoot trefoil or alfalfa early in pasture and hay regrowth, and avoided both species during the fall. Average SR values indicate selectivity for weed species was least on OG‐RC followed by OG‐BFT, OG, and OG‐ALF. Diets from grass‐legume systems contained similar amounts of weeds and legumes, but weed proportion increased with time.