RESUMO
Sequential feeding programs (Seq) of 2 feeds, 1 protein rich-energy poor and 1 energy rich-protein poor, during the same day or every other day were compared with a control complete diet for male broiler growth and body composition from 15 d to market weight. In experiment 1, BW gain and breast meat yield were significantly lower than those of controls for 24-h-cycle Seq with extreme protein content difference between the 2 feeds. BW gain and breast meat yield were higher than those of controls when feeds with moderate differences [feed moderately rich in protein (PM) = 26% CP; feed moderately rich in energy (EM) = 16% CP] were fed. Feeding periods that were half as long but changed twice as frequently gave less favorable results. In experiment 2, effects of allowing access to feed for different lengths of time were measured. The treatments in a 24-h cycle were a constant control diet, 50% PM-50% EM, 40% PM-60% EM, or 80% PM-20% EM decreasing to 40% PM-60% EM. Chickens fed 40% PM-60% EM had a 6% lower BW gain and a 3% higher feed/gain ratio and were fatter than those from other treatments. In experiment 3, male chickens fed a 48-h-cycle Seq of EM and PM every other day had the same BW gain, feed intake, and feed/gain ratio as controls. In experiment 4, five 48-h-cycle Seq with varying differences of energy and protein contents between EM and PM gave similar or slight reduction of performances compared with controls. A field trial with 8 flocks of broilers confirmed that feeding high and low protein feeds on alternate days resulted in performance similar to that from feeding a complete feed despite large day-to-day variations in lysine intake. Converse to shorter phases, Seq for 48-h cycles offers new opportunities for practical application and also for studies of short-term regulation of protein and lipid metabolism in chickens.