Wheat uniformity trials was conducted at experimental farm, Faculty of Agriculture, Kafrelshiekh university, during 2011/2012 winter season. The cultivar grown was Misr1, the experimental field consisted of 8 strips with 100 ridges in each strip. Thus, the basic unit was one row 0.2m wide and 3.0m long (area 0.6m²). Therefore, a total 800 basic units was used. Yield data recorded for the basic units were later combined to obtain the yield of different sized plot groupings. The data were subjected to two procedures of statistical analysis to estimate the optimum plot size, The first statistical method was the maximum curvature The second method was that developed by Smith (1938). Bartlett's test for homogeneity of variances, was used to study the effect of changing plot shape. The results obtained could be summarized as follows: Increasing the plot size decreased the variance per basic unit and the coefficient of variation. However, the reduction was not in proportion with the increase in plot size.Long and narrow plots were more efficient. The index of soil variability was 0.52 (intermediate soil heterogeneity). The optimum size of plot ranged from2 to 4 basic units (i.e.1.2 m² to 2.4m²).