RESUMEN
This study aimed to compare the digestibility of tropical grasses by horses by the in vivo method using mobile nylon bags with the in vitro digestibility method using horse feces as a source of inoculum. Five horses were used in a 2 × 5 factorial design with randomized blocks featuring two methods (in vivo and in vitro) and five grasses: Tifton 85 hay (Cynodon spp.), sixweeks threeawn grass (Aristida adsencionis, Linn), Alexandergrass (Brachiaria plantaginea (Link) Hitchc.), capim-de-raiz (Chloris orthonoton, Doell), and Sabi grass (Urochloa mosambicensis). No difference (P>0.05) was found between the in vivo and in vitro methods regarding nutrient digestibility of Sabi grass and sixweeks threeawn. Tifton 85 was the only grass that showed differences (P<0.05) between the two methods concerning the apparent digestibility of all nutrients. Alexandergrass, Tifton 85, and capim-de-raiz exhibited the best digestibility of dry matter, neutral detergent fiber, acid detergent fiber, and organic matter by the mobile bag method compared to the in vitro method. Tifton 85 and capim-de-raiz had higher crude protein digestibility by the mobile bag method than by the in vitro method. The mean retention time of the mobile bags in the digestive tract of the horses was 43.69 h. The bags with samples of sixweeks threeawn and Sabi grass had shorter retention times than capim-de-raiz and Alexandergrass (P<0.0001). It is concluded that, for sixweeks threeawn and Sabi grass, digestibility in horses can be assessed using the in vitro method in place of the mobile nylon bag method.
