Complementarities between grasses and forage legumes from temperate and subtropical areas on in vitro rumen fermentation characteristics

Associative effects between plants included in some mixtures with tropical and temperate species, which are able to grow in a wide range of temperature, deserve to be better investigated. The aim of this work was to assess the in vitro rumen fermentation characteristics when a tropical (peanut, Arachis pintoi) or temperate legume (sainfoin, Onobrychis viciifolia) with condensed tannins (CT), or a temperate legume without CT (alfalfa, Medicago saliva), were mixed with a tropical (axonopus, Axonopus catharinenses) or a temperate grass (tall fescue, Festuca arundinacea). Different proportions of each legume were mixed with each grass (in%, on dry matter (DM) basis, 0:100, 25:75, 50:50 75:25 and 100:0), and were tested using an in vitro rumen fermentation assay. The crude protein content in both tropical and temperate grasses ranged from 100 to 120 g/kg DM, but the aNDF content was much higher in the tropical (617 g/kg DM) than in the temperate grass (464 g/kg DM). The ADF (+ 214 g/kg DM) and lignin (+ 46.2 g/kg DM) content was much higher in the alfalfa when compared with the average of two other legumes. Positive quadratic effects were detected on in vitro DM disappearance (IVDMD), total gas production, and volatile fatty acids production, when alfalfa was mixed with axonopus, but similar response was not observed when alfalfa was mixed with tall fescue. The IVDMD and total gas production linearly increased when the proportion of peanut or sainfoin increased in the mixtures containing axonopus, but did not change when these legumes were mixed with tall fescue. The ammonia production increased when the proportion of legume increased in all mixtures. Reductions on ammonia and methane (CH4) productions were observed in mixtures containing sainfoin without the presence of polyethylene glycol (PEG) compared to fermentation done with PEG, but this response was not observed in the mixtures containing peanut. We conclude that complementarities in terms of energy and protein content can create favourable conditions leading to positive associative effects on rumen digestive parameters when legume species were mixed with axonopus, but not when they are mixed with tall fescue, probably due to different fibre content of grasses. The CT of sainfoin allow to reduce ruminal protein degradability and mitigate CH4 emissions per kg of DM, while the CT content of peanut is not enough to have the same effects.