Is goats' preference of forage trees affected by their tannin or fiber content when offered in cafeteria experiments?

Four forage trees were evaluated for chemical composition, in vitro apparent dry matter (DM) digestibility (IVDMD), in vitro gas production (IVGP), preference and intake rate (IR) by goats. Two cafeteria experiments were completed with four Criollo kids (17.1 +/- 1.39 kg live weight (LW)) in a repeated Latin square Design. In Experiment 1, fresh foliage of Lysiloma latisiliquum, Acacia pennatula, Piscidia piscipula, and Brosimun alicastrum were offered free choice for 4h. In Experiment 2, B. alicastrum was withdrawn from the experiment and preference for the remaining tree foliages was determinated. Daily samples were collected for chemical composition, and IVGP profiles were determined at 3, 6, 9, 12, 15, 18, 21 and 24 h of incubation with IVDMD measured at 24 h. Effective intake time and IR (g dry matter (DM)/min) was estimated from instantaneous measurements every minute during the 1 and 4 h. In the 4 h period of Experiment 1, B. alicastrum was preferred (13.56 g DM/kg LW), followed by P. piscipula (8.5 g DM), L. latisiliquum (8.46 g DM) and A. pennatula (5.18 g DM) (P<0.05). However, B. alicastrum had the lowest IR (5.9 g DM/min) (P<0.05). When B. alicastrum was withdrawn (Experiment 2), the same preference order was maintained but higher DM intake ocurred in the remaining plants: P. piscipula (11.77 g DM/kg LW), L. latisiliquum (10.28 g DM) and A. pennatula (7.23 g DM) (P<0.05). Nevertheless, no difference ocurred in the IR between plants. In Experiment 1, foliage DM intake was positively correlated with cellulose + hemicellulose (CEL + HEM) (r(2) = 0.69; P=0.003) and cellulose (r(2) = 0.63; P=0.008) and negatively correlated with condensed tannins (CT) (r(2) = -0.61; P = 0.01) and total tannins (TT) (r(2) = -0.57; P=0.02). In Experiment 2, foliage DM intake was positively correlated with hemicellulose (r(2) = 0.92; P<0.001) and CEL + HEM (r(2) = 0.75; P=0.01) and negatively with total polyphenols (r(2) = -0.74; P=0.02) and TT (r(2) = -0.73; P=0.02). Thus, in these short-term preference studies, fiber components were better predictors of DM intake than the polyphenolic compounds. If tannins are not involved in depressing DM intake of tanniniferous foliage, at levels typically found in them, then foliage with CT may have a possible role as a non-conventional anthelmintic for goats. (C) 2007 Elsevier B.V. All rights reserved.