A study on energy and protein utilization, and milk production of Bali cows on grass-legume diets was carried out using 12 first lactation cows (initial BW $263.79{\pm}21.66kg$) during a period of 16 weeks starting immediately post calving. The animals were randomly allotted into 4 dietary treatment groups R1, R2, R3 and R4, receiving from the last 2 months of pregnancy onwards, graded improved rations based on a mixture of locally available grass and legume feed ad libitum. R1 contained on a DM basis 70% elephant grass (PP, Penisetum purpureum) plus 30% Gliricidia sepia leaves (GS), R2 was 30% PP plus 55% GS supplemented with 15% Hibiscus tilliactus leaves (HT, defaunating effect), R3 and R4 were 22.5% PP+41.25% GS+11.25% HT+25% concentrate, where R3 was not and R4 supplemented with zinc di-acetate. TDN, CP and zinc contents of the diets were 58.2%, 12.05% and 18.3 mg/kg respectively for R1, 65.05%, 16.9% and 25.6 mg/kg respectively for R2, 66.03%, 16.71% and 29.02 mg/kg respectively for R3 and 66.03%, 16.71% and 60.47 mg/kg respectively for R4. Milk production and body weight were monitored throughout the experimental period. In vivo body composition by the urea space technique validated by the body density method and supported by carcass data was estimated at the start and termination of the experiment. Nutrient balance and rumen performance characteristics were measured during a balance trial of 7 days during the 3rd and 4th week of the lactation period. Results indicated that quality of ration caused improvement of ruminal total VFA concentration, increments being 52 to 65% for R2, R3 and R4 above R1, with increments of acetate being less (31 to 48%) and propionate being proportionally more in comparison to total VFA increments. Similarly, ammonia concentrations increased to 5.24 to 7.07 mM, equivalent to 7.34 to 9.90 mg $NH_3-N/100ml$ rumen fluid. Results also indicated that feed quality did not affect DE and ME intakes, and heat production (HP), but increased GE, UE, energy in milk and total retained energy (RE total) in body tissues and milk. Intake-, digestible- and catabolized-protein, and retained-protein in body tissues and milk (Rprot) were all elevated increasing the quality of ration. Similar results were obtained for milk yield and components with mean values reaching 2.085 kg/d (R4) versus 0.92 kg/d (R1) for milk yield, and 170.22 g/d (R4) vs 71.69 g/d (R1), 105.74 g/d (R4) vs 45.35 g/d (R1), 101.34 g/d (R4) vs 46.36 g/d (R1) for milk-fat, -protein, and -lactose, respectively. Relatively high yields of milk production was maintained longer for R4 as compared to the other treatment groups. There were no significant effects on body mass and components due to lactation. From the relationship $RE_{total}$ (MJ/d)=12.79-0.373 ME (MJ/d); (r=0.73), it was found that $ME_{m}=0.53MJ/kgW^{0.75}.d$. Requirement of energy to support the production of milk, ranging from 0.5 to 3.0 kg/d, follows the equation: Milk Prod. ($Q_{mp}$, kg/d)=[-2.48+4.31 ME($MJ/kg^{0.75}.d$)]; (r=0.6) or $Q_{mp}$=-3.4+[0.08($ME-RE_{body\;tissue}$)]MJ/d]; (r=0.94). The requirement for protein intake for maintenance ($IP_m$) equals $6.19 g/kg^{0.75}.d$ derived from the relationship RP=-47.4+0.12 IP; (r=0.74, n=9). Equation for protein requirement for lactation is $Q_{nl}$=[($Q_{mp}$)(% protein in milk)($I_{mp}$)]/100, where $Q_{nl}$ is g protein required for lactation, $Q_{mp}$ is daily milk yield, Bali cow's milk-protein content av. 5.04%, and $I_{mp}$ is metabolic increment for milk production ($ME_{lakt}/ME_{m}=1.46$).