Field experiments were conducted at the Rice Research Institute (Mazandaran Province) using randomized complete block designs in two fields of different levels of organic carbon (low and medium) during two cropping years. The treatments were repeated in three replicates and while seven nitrogen levels (Urea-N input) 0, 23, 46, 69, 92, 115 and 138 Kg N per ha employed. Results indicated that optimum grain yield was obtained from N92 (92 Kg N ha-1) in the low organic carbon containing field and from N115 (115 Kg N ha-1) in the medium organic carbon carrying field. Agronomic, recovery and physiological efficiency of the treatments with optimum grain yield in low vs. medium organic carbon fields were determined as 14.5 vs. 19.3 Kg of grain per Kg applied nitrogen, 37.6 vs. 35.9 percentage wise, and 36.8 vs. 50 Kg grain per Kg of absorbed nitrogen, respectively. The results revealed that the use of different levels of soil organic matter couldn’t be used as a suitable index for nitrogen fertilizer recommendation in wetland rice.