In two separate 2 x 2 factorial arrangements, 69 male, crossbred lambs were allotted randomly to the following treatments: 1) nonimplanted (NI) rams, 2) implanted (I) rams, 3) NI wethers, and 4) I wethers. In Trial 1, 36 lambs were allotted to treatment groups at birth (n = 9) and I lambs were implanted with 12 mg of zeranol between 1 and 3 d of age and again at weaning (average age of 62 d). Lambs were slaughtered at three time-constant end points of 78, 93, and 107 d on feed postweaning (average age of 155 d). Rams grew faster postweaning, were more efficient in their feed conversion, were heavier at slaughter, and had lower numerical yield grades than did wethers (P less than .05). Implanted lambs tended (P = .08) to grow faster and were (P less than .05) more efficient in their feed conversion than NI lambs. Rams produced heavier (P less than .05) trimmed subprimal shoulders, loins, and legs and had (P less than .05) a higher percentage of their carcass weight in the subprimal shoulder than did wethers. During Trial 2, NI rams (n = 8), I rams (n = 8), NI wethers (n = 8), and I wethers (n = 9) were allotted to treatment groups, and I lambs were implanted at average ages of 14, 55, and 98 d. After weaning, lambs were weighed every 14 d and were slaughtered 7 d after reaching a minimum weight of 50 kg (average age of 148 d).(ABSTRACT TRUNCATED AT 250 WORDS)