This article outlines the principles of radiobiology that can explain the time of onset, duration, and severity of the complex reactions of the lung to ionizing radiation. These reactions have been assayed biochemically, cell kinetically, physiologically, and pathologically. Clinical and experimental data are used to describe the acute and late reactions of the lung to both external and internal radiation including pneumonitis, fibrosis and carcinogenesis. Acute radiation pneumonitis, which can be fatal, develops in both humans and animals within 6 months of exposure to doses greater than or equal to 8 Gy of low LET radiation. It is divisible into a latent period lasting up to 4 weeks; an exudative phase (3-8 weeks) and with an acute pneumonitic phase between 2 and 6 months. The latter is an inflammatory reaction with intra-alveolar and septal edema accompanied by epithelial and endothelial desquamation. The critical role of type II pneumonocytes is discussed. One favored hypothesis suggests that the primary response of the lung is an increase in microvascular permeability. The plasma proteins overwhelm the lymphatic and other drainage mechanisms and this elicits the secondary response of type II cell hyperplasia. This, in its turn, produces an excess of surfactant that ultimately causes the fall in compliance, abnormal gas exchange values, and even respiratory failure. The inflammatory early reaction may progress to chronic fibrosis. There is much evidence to suggest that pneumonitis is an epithelial reaction and some evidence to suggest that this early damage may not be predictive of late fibrosis. However, despite detailed work on collagen metabolism, the pathogenesis of radiation fibrosis remains unknown. The data on radiation-induced pulmonary cancer, both in man and experimental animals from both external and internal irradiation following the inhalation of both soluble and insoluble alpha and beta emitting radionuclides are reviewed. Emphasis is placed on the data showing that alpha emitters are at least an order of magnitude more hazardous than beta/gamma radiation and on recent data showing that the more homogeneous the irradiation of the lung, the greater is the carcinogenic hazard which contradicts the so-called "hot particle" theory.