Man-made grasslands dominated by Imperata cylindrica (L.) Beauv. in forested areas of lowland Nepal are commonly cut and/or burned annually. Changes in grass forage quality following different treatments of cutting and burning and axis deer (Axis axis) response to such habitat manipulations were investigated. Samples of matured grass were collected in December 1990, February and April 1991 from three experimental sites: cut, burned, cut-and-burned. Four locations on cut-and-burned grassland were repeatedly sampled at 12-d intervals from January to April 1992. Numbers of axis deer were recorded during the dry season of 1991/1992 on grassland plots receiving the following treatments: cut, cut-and-burned, and uncut/unburned (controls). Based on grass quality differences between December and February and between December and April, cut-and-burned treatments gave the greater increase in forage quality. N was significantly higher on cut-and-burned plots than on cut plots both in February and in April, while Na, K and P was significantly higher in February. On plots cut-and-burned in January, Ca concentrations were relatively low while the P content fell below required levels for domestic stock towards the end of the dry season in April. Na concentrations were below the minimum required levels for both domestic and wild ruminants during the whole period. When an entire grassland was cut, deer density increased gradually. When the same area was subsequently burned, the daily deer density increased much more rapidly. Axis deer preferred burned plots compared to plots neither cut nor burned and to cut plots. Plots burned in late February had higher densities of axis deer than plots burned 1.5 mo earlier. When nearby recently burned plots were available, deer density was reduced on plots burned earlier.