Theoretical expressions have been derived for the contributions to ionisation density in cavity chambers from secondary electrons arising from photon absorption in wall, build-up cap and surrounding medium. The method allows the derivation of C lambda and displacement factors, which are evaluated for the Farmer graphite chamber for a range of photon beams. Differences between C lambda values calculated for air-walled and water-walled chambers are found to be consistent with those determined by Nahum and Greening. The currently recommended displacement factor for the Farmer chamber with cap, however, is found to be in error by about 0.8%. On the basis of a comparison between theoretically predicted air doses in cavities and the experimental results of Fregene it is proposed that no further correction to C lambda is required for the chamber 'effective point of measurement'.