INTERRELATIONSHIPS BETWEEN CHEMICAL-CONSTITUENTS, RUMEN DRY-MATTER AND NITROGEN DEGRADABILITY IN FRESH LEAVES OF MULTIPURPOSE TREES

This work determined the inter-relationships between chemical constitu ents, rumen dry matter (DM) and nitrogen (N) degradabilities of fresh leaves from 20 multipurpose tree (MPTs) accessions belonging to the fo llowing genera: Acacia (2), Cajanus (2), Chamaecytisus (2), Erythrina (2), Leucaena (8) and Sesbania (4). Significant differences (P < 0.05) were observed among genera in all assayed chemical attributes except hemicellulose. There was apparently no disappearance of N in A saligna after the initial solubility. All genera had similar DM and N solubil ity but varied in degradation rate of DM (0.064-0.137 h(-1); P < 0.002 ) and N (0.062-0.135 h(-1); P < 0.001), effective degradability of DM (672-856 g kg(-1); P < .001) and effective degradability of N (NED; 70 1-902 g kg(-1); P < 0.001). Except for A saligna, synchronization indi ces of the release of nutrients for each accession ranged from perfect (0.93) to poor (-0.31) for soluble nutrients and from moderate (0.75) to very poor (-0.81) for insoluble slowly degradable nutrients becaus e N was in excess. Inter-relationships among chemical constituents, DM and N degradation constants were established using correlation and re gression analyses. Negative correlations occurred between N solubility and proanthocyanidins (PAs; P < 0.05), N degradation rate and hemicel lulose (P<0.05), NED and either NDF (P < 0.05) or hemicellulose (P < 0 .05) or PAs (P < 0.01), while N degradation rate was positively correl ated with N content (P < 0.01). Equations for predicting N degradabili ty characteristics (solubility, insoluble slowly degradable fraction, potential and effective degradability, rate of degradation and lag pha se) accounted for 0.44-0.83 of the variation (using chemical constitue nts), 0.56-0.81 of the variation using corresponding DM degradation ch aracteristics and, 0.74-0.89 of the variation using mixed criteria. Su perior models were of the form: ND_P = DM_P(-beta), where ND_P and DM_ P are corresponding N and DM degradability characteristics, respective ly, and beta the regression of the deviations of the ratio ND_P : DM_P from unity on chemical constituents and accounted for 0.75-0.92 Pract ical implications of DM degradability and nutrient release synchrony i ndices in relation to basal roughage intake and microbial protein synt hesis were discussed. It was concluded that a mixed model could adequa tely predict N degradability of fresh forages and alleviate analytical cost. A functional approach of classifying browses is outlined.