Selection in three lines of Hereford cattle for 1) weaning weight (WWL), 2) yearling weight (YWL), and 3) an index of yearling weight and muscle score (IXL) was studied. Remnant foundation cows and semen from seven foundation sires were used to establish an unselected control line for the last 11 yr of the experiment. Performance data collected over a 23-yr period on birth weight (BWT), weaning weight (WWT), postweaning gain (PWG), yearling weight (YWT), muscle score (MSC), and an index (IDX) giving equal weight to standard deviations of yearling weight and muscle score were analyzed. Generation interval of midparents was about 4.16 yr in each selected line. Sire and dam selection differentials, in standard deviation units per generation, for primary criteria were, respectively, 1.59 and .33 for WWT in WWL, 1.75 and .25 for YWT in YWL, and 1.42 and .25 for IDX in IXL. Components of direct and maternal genetic variances, direct-maternal covariance, and dam permanent environmental variance were estimated by REML. The average annual response of males and females in actual units for each trait in WWL, YWL, and IXL was, respectively, BWT, .22, .24, and .27 kg; WWT, .98, .63, and 1.26 kg; YWT, 2.43, 2.64, and 3.44 kg; and MSC, .053, .009, and .104 scores. Average selection responses in BWT, WWT, YWT, MSC, and IDX per unit of primary criteria in each selection line (all in standard deviation units) were .22, .20, .31, .10, and .24 for WWT in WWL; .23, .12, .32, .04, and .21 for YWT in YWL; and .27, .22, .40, .20, and .36 for IDX in IXL. Responses in bold type are realized heritability and others are correlated responses. Realized genetic correlations were .78 for WWT and YWT, .87 for WWT and IDX, and .86 for YWT and IDX. Responses for all traits in IXL were greater than in other selected lines.