Additive series of  experiments were conducted under field conditions to detect the effect of duration of interference of different initial population densities(5,10,20and 30plants/m2) of purple nutsedges (Cyprus rotundus L.)on the yield losses of drill-seeded rice(Giza 178). Increasing weed population or/and the duration of interference increased reduction of rice grain yield. The severe reduction  due to weed population of 30 plants/m2 recorded 31and 27% for the first and second seasons, respectively. The highest reduction due to duration of interference recorded 53 and 39% when purple nutsedge was allowed to compete for 120 days with rice during first and second seasons, respectively. Moreover, the interaction between weed densities and duration of interference showed severe reductions under the high populations(20 and 30 plants/m2) when interfered with rice 40 days or more