Roentgen-calibrated cavity-ionization chambers are frequently used to determine absorbed dose at a point in a tissue-like phantom. An N.P.L. calibration factor for 2 MV X-rays, assigned to a chamber and its associated measuring equipment, includes corrections for wall attenuation of the chamber, for any lack of electronic equilibrium, and for the stopping power of the wall material relative to air. When the chamber readings are used to calculate absorbed dose at a point in, say, a water phantom, a further correction is required to determine the exposure dose that would have been obtained at the point in the absence of the chamber and any wall-thickening cap used in the calibration. Some recent work dealing with the determination of this correction has been critically examined. It is shown that the correction factor required to compensate for the attenuation of the phantom material removed to accommodate the chamber and its cap can be determined from experimental data already available. A rad/roentgen conversion factor for 2 MV X-rays and a water phantom is calculated.