Introduction: Skin sparing is one of the most desirable characteristics of high energy photon beams. However, the photons emerging from the target of linacs are contaminated by secondary electrons as a result of their interactions with air, collimators, flattening filter and any other objects in their path. This phenomenon tends to increase the skin dose received by the patients. A practical and simple way to reduce the contribution of electron contamination is to place a sheet of medium to high Z material just after the secondary collimator. In this study, filters having different thickness and atomic number were applied and their effectiveness on the reduction of skin dose was evaluated. Materials and Methods: The filters of different thickness and atomic number were applied. The percent depth dose values were determined by the direct measurements made in a Scanditronix water phantom using a PTW 31006 Pin Point chamber having a sensitive volume of 0.015 cm 3 . A Perspex filter holder was made to be installed on the accessory slot. A plastic bag containing helium was also made using thin plastic sheet to study the effect of the helium bag when it replaces the air column between the head of the linac and the phantom. All of the measurements were carried out for the three field sizes of 10×10, 20×20 and 25×25 cm 2 . The setups were adjusted for SSD = 100 cm. The ratio of the surface dose to maximum dose (Ds) was used as the criterion to determine the optimum filter. Results: The dosimetry results obtained in the water phantom indicated that a 0.4 mm thick Pb filter is the most effective one. This filter reduces the Ds for the field sizes of 10×10, 20×20 and 25×25 cm 2 by 5.7, 7.9 and 9.6%, respectively. Also the simultaneous use of the optimum filter and He bag is more effective than the filter alone. It reduces the Ds by 6.3, 10.1 and 12.3% for the field sizes of 10×10, 20×20 and 25×25 cm 2 , respectively. Discussion and Conclusion: Based on the results of this work it is evident that the contribution of contaminant electrons to dose from the air column between the head and the phantom is much smaller than it from the secondary electrons arising from the head of the linac. On the other hand, the electron contamination originating from the air column is almost independent of the field size. But the surface dose arising from the secondary electrons produced by the head of the linac depends on the field size, which is increased by increasing the field size.