Abstract
Four, mature rumen fistulated male swamp buffaloes with an average initial weight of $426{\pm}25kg$ were randomly allocated to receive dietary treatments according to a $4{\times}4$Latin square design. Four dietary treatments with varying proportions of ureatreated rice straw (UTRS) and cassava hay (CH) were offered (100:0, 75:25, 50:50, 0:100). Each feeding period lasted 21 days, the first 14 days for feed adjustment and intake measurement and the final 7 days for rumen fluid and faecal collections. The results demonstrated the potential use of both UTRS and CH as roughage sources. As levels of CH increased in the diets pH values were maintained (6.5-7.0). Ruminal $NH_3-N$ concentrations were significantly (p<0.05) raised as higher levels of CH were incorporated into the diets. Moreover, cellulolytic and proteolytic bacterial populations were enhanced while total protozoal counts were decreased (p<0.05). In addition, DM, OM and CP digestibilities and their digestible intakes as well as estimated energy increased with increasing levels of CH in the diets. The results suggest a favorable effect of CH substituting for UTRS.