Nutrient composition and in vitro ruminal fermentation of tropical legume mixtures with contrasting tannin contents

Various combinations of a low-tannin herbaceous legume (Vigna unguiculata) and foliage of tanniniferous shrub legumes (Calliandra calothyrsus, Flemingia macrophylla and Leucaena leucocephala) or a low-tannin shrub legume (Cratylia argentea), all mixed together with a low-quality tropical grass (Brachiaria humidicola), were tested in vitro for differences in the effects on ruminal fermentation. Two experiments with the gas transducer technique were carried out, where each forage mixture was tested either with or without polyethylene glycol in order to be able to identify tannin-related effects (n = 3). In Experiment 1, a stepwise replacement of V. unguiculata by C. calothyrsus (5:0, 4:1, 3:2, 2:3, 1:4, 0:5) at a legume proportion of 1/3 or 2/3 in the mixture was evaluated. Together with two grass-alone and four pure legume treatments this added up to 30 treatments. In Experiment 2, V. unguiculata was gradually replaced by each of the four shrub legumes (3:0, 2:1, 1:2, 0:3) in grass–legume ratios of 2:1, adding up, together with two grass-alone treatments, to 28 treatments. When added alone, V. unguiculata resulted in high fermentative activity as measured by gas production and kinetics as well as low proportion of undegraded crude protein. When V. unguiculata was replaced by the low-tannin C. argentea in Experiment 2, there was no noticeable difference (P>0.05) in fermentative activity. In both experiments, the effect of the substitution of V. unguiculata by tanniniferous shrub legumes resulted in a declining gas production and an increasing proportion of undegraded crude protein (P<0.001). However, the extent of these changes depended on the level of replacement and the shrub legume species (P<0.001). The results of Experiment 2 illustrate that this was the consequence not only of different tannin contents (less adverse effects with L. leucocephala than with C. calothyrsus) but also differences in the chemical properties of the tannins present in these shrub legume species (much less adverse effects with L. leucocephala than with F. macrophylla despite similar tannin contents). Furthermore these results indicate that, once the extent of the effects of a tanniniferous legume is known, one may calculate the maximal level of replacement of a low-tannin legume in a grass diet possible without negative effects on ruminal fermentation. This allows to improve dry season grass-based diets with as few as possible of the expensive and less well growing low-tannin legume.