Selection for eight-week body weight in two randombred chicken populations under altered water to feed ratios. 1. Selection responses.

A study was conducted to investigate the response to selection for 8-wk BW under selection environments that utilized different dietary salt levels to alter water to feed ratios. The Athens-Canadian (AC) and Athens Randombreds (ARB) served as base populations. A broiler diet consisting of 23% CP and 3,120 kcal of ME/kg was modified to contain .2, .4, and 1.6% dietary salt. Chicks from the AC and ARB populations were divided within full-sib families into three groups, with one group assigned to each of the three dietary salt selection environments. Individual phenotypic selection was conducted within the six lines for high 8-wk BW for six generations with unselected randombred controls included in each generation to measure environmental variation. No differences in 8-wk BW were observed among the three salt selection environments. Lines derived from the ARB population were initially approximately 75 g larger than those derived from the AC population. However, when BW were expressed as percentage deviation from the control in the sixth generation, selection response patterns were similar (AC lines, 28 to 32%; ARB lines, 32 to 36%). These data indicated that selection for 8-wk BW was as effective under high- and low-salt diets as under a normal-salt diet. Realized heritabilities, following correction for environmental variation, ranged from .38 to .46. Failure to correct for environmental variation resulted in large fluctuations in heritabilities across generations. These data provide strong support for inclusion of controls to accurately measure the heritability of BW in selection studies.