A comparison of methods for measuring the nutritive value for pigs of a range of hulled and hulless barley cultivars

Six hulled and three hulless barley cultivars were each divided into a high and a low density fraction by the use of a gravity table. These fractions were analyzed for density (wt vol−1), 1000 kernel weight, soluble non-starch polysaccharides (S-NSP), insoluble NSP (I-NSP), lignin, total dietary fibre (TDF, S-NSP plus I-NSP plus lignin), β-glucans, gross energy and crude protein content. Measurements were made on pigs and rats for apparent energy digestibility (AED) and true protein digestibility (TPD, apparent digestibility minus metabolic faecal N), apparent dry matter digestibility in pigs by the mobile nylon bag (MNB) technique, and in vitro digestibility of dry matter and N by simulation of enzymatic activity anterior to the terminal ileum and for the whole gastrointestinal tract. The results obtained were examined for the reliability of various measurements for predicting AED and TPD in pigs. Coefficients of determination were assessed for all sets of measurements for both the total 18 samples and for the 12 hulled samples only. For most measurements, the predictive values were greater when 18 samples were used than when 12 samples were used. Best predictions for AED in pigs were AED in rats (r2, 0.97, n = 18; 0.88, n = 12), I-NSP (r2, 0.97, n = 18; 0.88, n = 12), in vitro organic matter digestibility (r2, 0.97, n = 18; 0.82, n = 12) and MNB dry matter digestibility (r2, 0.86, n = 18; 0.85, n = 12). Coefficients of determination were not so high for the estimation of TPD in pigs. The only values showing a strong relationship were obtained for the TPD measurements in rats (r2, 0.94, n = 18; 0.88, n = 12).