In recent years there has been a growing interest in all forms of rotational or moving-field therapy, and many different types of therapy machines designed for this kind of treatment have become available. To the medical radiation physicist, the dosimetry of rotation therapy has presented a number of interesting problems, and much useful work has been published on the basic data of dose distribution and dosage calculation. All of the currently popular methods of dose determination—except direct measurement on the patient, which, for good reasons is rarely popular—assume that the absorbing medium is homogeneous and water-equivalent. For many purposes, this may be adequate, but for irradiation of certain parts of the human body, such as the pelvis or the thorax, it is quite apparent that this assumption breaks down badly, especially at conventional x-ray energies. Even with cobalt-60 radiation, discrepancies of 20 to 30 per cent between calculated and measured doses are found in irradiating such heterogeneo...