The manning roughness coefficient n is one of the parameters in designing furrow irrigation. Its appropriate selection is important for efficient water application in field. In this research, the Manning roughness coefficient in ordinary and every other furrow irrigations with different inflow rates (0·4, 0·8, 1·1 l s −1) and furrow slopes (0·2 and 0·4%) at various growth stages of wheat with different crop vegetation covers were determined on a clay loam soil. The calculation was based on the volume balance equation in the form of a differential equation that was solved with the forward finite difference (method I) and the secondary backward finite difference (method II) procedures. Comparison of the results obtained from these two methods showed that method I resulted in lower values of n than method II. So, the mean of the values for n from the two methods were used for the medium texture soil of the present study. There was no difference between the values of n in ordinary and every-other furrow irrigations. The results also indicated that at the first irrigation, the values of n were high (0·07–0·121), but at the second and third irrigations, the values of n decreased by about 60–70%. However, after the third irrigation, the seedling emergence of wheat resulted in increasing the values of n. The minimum and maximum values of n which belonged to the third and seventh irrigations were 0·047 and 0·136, respectively. Finally, two equations were proposed for the estimation of the roughness coefficient for bare and vegetated furrows according to the number of irrigations (up to three before seedling emergence), inflow rates and percentage of vegetation covers.