Tall fescue (Festuca arundinacea Schreb.) infected with the non-toxic novel endophyte (Neotyphodium coenophialum) may have potential in areas south of the traditional tall fescue zone in the USA. A 3-yr study conducted at Raymond, MS compared forage and animal responses on novel endophyte-infected ‘GA5' and ‘Jesup', or endophyte-free ‘Kentucky 31' tall fescue, and ‘Marshall' ryegrass (Lolium multiflorum Lam.) pastures in a completely randomized design experiment with two replications. Beef steers (initial BW = 240 kg; stocking rate = 3 steer ha^-1) grazed 1.6-ha experimental units for 24 wk starting in early or mid December. Year 1 results indicated that average daily gain (ADG) of steers was similar among the tall fescue treatments (0.91 kg), which was lower than on ryegrass pastures (1.15 kg). Year 2 ADG was generally 0.23 kg less than for Year 1, but magnitude of difference between forage species was the same. Year 3 ADG on ryegrass (1.18 kg) was still greater than on tall fescue, but among tall fescue treatments, ADG was similar on GA5 (0.91 kg) and Jesup (0.86 kg), but was lower on Kentucky 31 (0.77 kg). In the first two years, average herbage mass (HM) was not different among tall fescue (2.2 - 2.4 Mg ha^-1 in Year 1; 2.4 - 2.7 Mg ha^-1 in Year 2), but was lower on ryegrass pastures (1.6 and 1.5 Mg ha^-1). In Year 3, however, tall fescue HM was less than ryegrass (1.8 Mg ha^-1), and was lower on Kentucky 31 (1.2 Mg ha^-1) compared to GA5 and Jesup pastures (1.3 Mg ha^-1 each). These results indicate that acceptable ADG may be obtained from non-toxic endophyte-infected tall fescue, but its suitability in warmer zones will depend on long-term pasture persistence and subsequent economic coefficients.