The absorbed dose, the mean value of the energy imparted by ionizing radiation to a volume of interest divided by the mass of that volume, is often a very effective way to describe radiation exposures. For example, in radiation therapy and industrial radiation processing, the absorbed dose is usually adequate for predicting the results of the irradiation. However, in situations in which the heterogeneity of the energy deposition or of the target structure results in individual targets (often assumed to be a cell or a cell nucleus) receiving energy depositions that are dramatically different from the mean value, the absorbed dose does not provide sufficient information needed to fully understand the consequences of irradiation or for the use of radiation-response models. In some cases, such as microbeam irradiations, heterogeneous exposures have been created intentionally in order to study mechanisms of response to ionizing radiation. In other situations, such as background-radiation exposure, the heterogeneity occurs as a result of the low level of radiation exposure or of a small local concentration of a radioactive material. When the energy deposition is heterogeneous, the conventional description in terms of absorbed dose can be misleading because it suggests that neighboring structures will incur the same amount of damage, and that the amount received by an individual target can be decreased to as low a value as desired. In fact, the individual targets can receive highly variable energy depositions, with a mean and variance determined by the physical properties of the radiation and the target. Reliance on the mean values rather than the spectrum of individual energy depositions can lead to inappropriate conclusions about the relationship between energy deposited and the initiation of biological response. Even with the same amount of deposited energy, the spectrum of initial damage products depends on radiation quality (sometimes specified in terms of the linear energy transfer). However, the complex combinations of biomolecular processes occurring following irradiation make it unlikely that any single-parameter description of the radiation interaction will be satisfactory for understanding biological processes. This Report recommends that, in cases in which energy deposition is likely to be heterogeneous, a detailed description of the radiation field, with an appropriate description of the irradiated system, should be given. The complete description of the radiation is the energy distribution of the particle radiance as a function of time and particle type. However, simplified descriptions such as the distribution of fluence rate, the probability density of lineal energy, and the event rate, or even the absorbed dose rate and radiation quality are appropriate descriptions of the irradiation in many cases.