Hard, dynamic or virtual wedges are often used in conformal radiation therapy to reduce dose inhomogeneity within the target volume. There are restrictions exist in using such field modifiers for large fields due to technical peculiarities of the treatment machine. Field-in-field (FIF) manual segmentation, where one or more subfields are used to achieve dose homogeneity, could solve this problem, but such technique needs to be evaluated and verified first. FIF technique is based on intensity modulated radiation therapy (IMRT) principle, except that the intensity modulation is done manually using direct planning. The aim of this article is to determine whether the gamma criterion evaluation principle is suitable for FIF plans verification. For this study a 17 x 12 cm rectangular 6MV photon beam was used. Rectangular fields, sized 5 x 12, 6 x 8,5 and 3 x 4,3 cm were added as segments for each 17 x 12 cm base field in center. Dose distribution was calculated using treatment-planning system (TPS) to be used as a starting point to make distorted plans with segments displaced in superior, inferior, lateral left and lateral right directions to simulate patient movement or positioning errors during irradiation. Dose distributions for the distorted plans were measured using pixel ionization chambers detector array. The resulting dose distributions were compared to the reference one provided by TPS using gamma criterion. For acceptance criteria ∆DM=3% dose-difference and ∆dM=3 mm distanceto- agreement (DTA) were used and 95% of all pixels should be within this criterion. For some plans, considered in the present paper, 95% threshold was not exceeded even when the displacement reached clinically significant values of 1 cm and even more. Thus, one have to conclude that use of gamma criterion with 95% threshold of number of pixels in agreement is not suitable for FIF plans.