The head scatter dose contribution to the output of a treatment machine has been determined for an open and wedged 60Co gamma-ray beam and for open and wedged x-ray beams of 4, 8, and 16 MV. From those data wedge factor values 'in air' have been deduced, expressed as the ratio of the dose to water, measured in air, for the situation with and without wedge, for the same number of monitor units (or treatment time for 60Co). The measurements have been performed using a polymethylmetacrylate (PMMA) and a graphite- wailed ionization chamber inserted in a brass build-up cap and in a PMMA mini-phantom, respectively. Absolute wedge factor values deduced with both detector systems and based on the ratio of ionization chamber readings, differ for the investigated photon beams, up to 3.5% for the 4 MV x-ray beam. The deviations result from the difference in composition between the detector materials and water and can be taken into account by conversion of the ionization chamber readings for both the open and wedged photon beams to the absorbed dose to water. For the brass build-up cap detector system the ratio of the conversion factors for the wedged and open beam changes the ratio of the ionization chamber readings up to about 3.6% for the 4 MV x-ray beam. For the mini-phantom the conversion factors for the wedged and open beam are almost equal for all photon beams. Consequently, for that system wedge factors based on ionization chamber readings or dose values are the same. With respect to the wedge factor variation with field size a somewhat larger increase has been determined for the 60Co and 4 MV photon beam using the brass build-up cap: about 1% for field sizes varying between 5 cm X 5 cm and 15 cmX 15 cm. This effect has to be related to an apparent more pronounced variation of the head scatter dose contribution with field size for the wedged photon beams if the brass build-up cap detection system is used. It can be concluded that determination of wedge factors 'in air' under reference irradiation conditions, performed with both the mini-phantom and brass build- up cap yields within 0.5% the same result if the wedge factors are based on a dose to water ratio. However, by using high-Z build-up materials the determination is more complicated because appropriate conversion factors are then required, while similar conversion factors can be ignored if more water equivalent build-up materials such as PMMA are applied.