The effectiveness of prescribed fire as a range management tool was examined in this study in the Yazd province of Iran. One hectare of land was divided into four 20 x 125 m blocks. Blocks were split into five 20 x 25 m experimental plots. Seasonal burning was randomly applied on each plot in 2004. One plot was also left unburnt as a control. Data were collected in 2005 and 2006 and analyzed. The results show that the response of species to fire varies within the season as well as among seasons. The minimum burning effect on vegetation cover was observed for the most desirable plant (Salsola rigida) in response to winter fire treatment. Fire eliminated both Artemisia sieberi and Stipa barbata, the two other important species found on the range. Reduction of undesirable species, such as Noaea mucronataand Cousinia deserti was also considerable. The percentage cover of other undesirable species, i.e. Scariola orientalis, did not change significantly (p < 0.05). We concluded that, although it burning eliminated some species, it could not eradicate all undesirable species. It also failed to promote the desired species in terms of vegetation cover. So, at least in the short term, burning is not considered a range improvement tool for steppe rangelands in Iran.