The objectives of this experiment were to determine the effects of replacing ground corn with soybean hulls (SH) in high-concentrate diets on the growth (56-d period), carcass characteristics, and eating behavior of feedlot lambs. Sixty-four Santa Inês ram lambs (18.3 ± 2.8 kg of BW and 69 ± 5 d of age) were assigned to a randomized complete block design experiment with 8 blocks and 4 diets. The control diet contained 10% coastcross (Cynodon sp.) hay, 70% corn, and no SH (SH0) in the dietary DM. In the remaining diets, SH replaced corn at the rate of 15 (SH15), 30 (SH30), or 45% (SH45) of the original corn concentration, resulting in 0, 10.5, 21.0, or 31.4% SH in the dietary DM. Dry matter intake increased linearly (P < 0.01) when SH replaced ground corn (1.0, 1.0, 1.1, and 1.1 kg/d for SH0, SH15, SH30, and SH45, respectively). There was no effect on ADG of lambs, with values of 276, 278, 282, and 287 g for SH0, SH15, SH30, and SH45, respectively. Feed efficiency decreased linearly (P < 0.01) with SH inclusion. Carcass measures were not affected by SH as a replacement for ground corn. Eating time, expressed as minutes per day and minutes per gram of NDF, showed a quadratic effect (P < 0.05), whereas no effect was observed when expressed as minutes per gram of DM. Rumination, in minutes per day, was not influenced by dietary SH inclusion, but a linear decrease (P < 0.01) was observed when this variable was expressed as minutes per gram of NDF. Soybean hulls can replace up to 45% of the ground corn (31.4% of SH in the dietary DM) in high-concentrate diets fed to feedlot lambs without negative effects on ADG and carcass measures. The linear decrease in feed efficiency (11.6% reduction from SH0 to SH45) suggests that optimal dietary SH inclusion rates should be dictated by the relative costs of SH and corn.