Forty-eight medium-framed, crossbred steers (average BW = 268 +/- 20 kg) were limit-fed a 35% concentrate diet to gain approximately .10 kg/d during a 66-d restriction (REST) period. This was followed by a 98-d realimentation (REAL) period in which steers were allotted to a 3 x 2 factorial arrangement of treatments. Treatments were 80% concentrate diets containing 9, 12, or 15% CP, and implantation of one-half of the steers with 24 mg/steer of estradiol-17 beta (E(2); Compudose(R)). The allotted steers were distributed across four pens (12 steers/pen) with each pen containing two replicates of each treatment combination. Individual feed intakes were monitored using a Calan(R) gate system. Each steer had ad libitum access to its respective diet. Longissimus muscle area and fat thickness at the 12th rib were measured using real-time ultrasound. Dry matter intake was not affected by CP or E(2) through d 56. Steers fed 9% CP weighed less (quadratic, P < .05) after d 14 than steers fed 12 or 15% CP due to slower (quadratic, P < .05) gains during this initial period. Implantation increased (P < .05) ADG from d 15 to 56 resulting in heavier BW (P < .05) on d 56 and 98. Increasing concentrations of dietary CP resulted in linear(P < .05) increases in longissimus muscle area on d 28 and 56, and fat thickness on d 98. Serum IGF-I concentrations were lowest on d -21 of REST and highest on d 28 of REAL (quadratic, P < .05). Implantation increased (P < .01) IGF-I throughout REAL, whereas IGF-I concentrations were increased (quadratic, P < .05) in steers fed 12 and 15% CP on d 28 only. Increasing concentrations of dietary CP enhanced BW gain and serum IGF-I in cattle during the initial phase of REAL, whereas E(2) affected later phases. These early changes affected BW and carcass characteristics during the remainder of REAL.