The objective of this study was to test whether the use of tannin-rich shrub legume forage is advantageous for methane mitigation and metabolic protein supply at unchanged energy supply when supplemented in combination with tannin-free legumes to sheep. In a 6 × 6 Latin-square design, foliage of two tannin-rich shrub legume species (Calliandra calothyrsus and Flemingia macrophylla) were used to replace either 1/3 or 2/3, respectively, of a herbaceous high-quality legume (Vigna unguiculata) in a diet composed of the tropical grass Brachiaria brizantha and Vigna in a ratio of 0.55 : 0.45. A Brachiaria-only diet served as the negative control. Each experimental period lasted for 28 days, with week 3 serving for balance measurement and data collection inclusive of a 2-day stay of the sheep in open-circuit respiration chambers for measurement of gaseous exchange. While Vigna supplementation improved protein and energy utilisation, the response to the partial replacement with tannin-rich legumes was less clear. The apparent total tract digestibilities of organic matter, NDF and ADF were reduced when the tannin-rich plants partially replaced Vigna, and the dose–response relationships were mainly linear. The tannin-rich plants caused the expected redistribution of more faecal N in relation to urinary N. While Flemingia addition still led to a net body N retention, even when fed at the higher proportion, adding higher amounts of Calliandra resulted in body protein mobilisation in the growing lambs. With respect to energy, supplementation of Vigna alone improved utilisation, while this effect was absent when a tannin-rich plant was added. The inclusion of the tannin-rich plants reduced methane emission per day and per unit of feed and energy intake by up to 24% relative to the Vigna-only-supplemented diet, but this seems to have been mostly the result of a reduced organic matter and fibre digestion. In conclusion, Calliandra seems less apt as protein supplement for ruminants while Flemingia could partially replace a high-quality legume in tropical livestock systems. However, methane mitigation would be small due to associated reductions in N and energy retention.

The effects of supplementing a tropical, low-quality grass hay (Brachiaria dictyoneura) with legume foliage (Cratylia argentea) or fruits of the multipurpose tree Sapindus saponaria on ruminal fermentation, methane release and nitrogen (N) utilization were evaluated. Six Swiss White Hill lambs were used in a 6 ✕ 6 Latin-square design with a 3 ✕ 2 factorial arrangement of treatments with measurements of energy metabolism being conducted using open-circuit respiratory chambers. Treatments consisted of three basal diets, either grass alone or legume: grass ratios of 1: 2 or 2: 1. These basal diets were supplemented (1: 3) with a control concentrate or with a concentrate containing 250 g/kg dry matter of S. saponaria fruits. The apparent total tract digestibilities of organic matter (OM) and neutral-detergent fibre (NDF) were reduced and the proportionate crude protein (CP) losses through faeces were increased (P 0·01) by supplementation with S. saponaria, and digestibilities of OM and NDF were linearly reduced (P 0·001) with increasing legume proportion. Body energy retention, however, was similar in all diets. Along with CP intake, the proportionate CP losses through faeces decreased (P 0·001) with increasing legume proportion which was associated with improved (P 0·001) body protein retention and reduced (P 0·1) fat retention. Ruminal fluid ammonia concentration was not significantly affected (P > 0·1) by the inclusion of S. saponaria in the concentrate, but increased linearly (P 0·001) as dietary legume proportion was elevated. Supplementation with fruits of S. saponaria increased (P 0·01) total bacteria count, and decreased (P 0·001) total ciliate protozoa count by more than proportionately 0·50. Daily methane release was reduced (P 0·01) by S. saponaria supplementation in all basal diet types. Although being not clearly affected on a daily basis, methane release relative to body protein retention decreased linearly (P 0·05) with increasing legume proportion. The fact that interactions were mostly non-significant (P > 0·05) indicates that supplementation with S. saponaria fruits is a useful means to reduce methane emission from sheep given both tropical grass-based and grass-legume-based diets. Likewise, including legumes in N-limited tropical diets seems to represent an environmentally friendly way to improve animal productivity.