DNA damage appears to be ubiquitous in the biological world, as judged by the variety of organisms which have evolved DNA-repair systems. Previously, it was proposed that germ-line DNA of multicellular organisms may be protected from damage, and consequently from aging, by efficient recombinational repair during meiosis. The somatic line, however, may be vulnerable to the accumulation of DNA damage, and hence undergo aging, owing to relatively less repair. Although the DNA lesions most important in aging are not known yet, there is evidence for serveral types of endogenous damage. DNA lesions have been shown to interfere with transcription and replication, and so lead to loss of cell function and death. In mammals, there is a progressive decline of function in many different tissues with increasing age. Deterioration of central nervous system functions appears to be a critical part of the aging process. This may be due to the low DNA repair capacity which is found in postmitotic brain tissue, and which could result in the accumulation of DNA lesions in this tissue. Also reviewed is evidence that species longevity is directly related to tissue DNA-repair capacity and that aging may be accelerated by treatment with DNA-damaging agents, or in individuals with genetically defective repair. Although it has been frequently postulated that somatic mutation may be cause of aging, current evidence suggests that it is probably less important than DNA damage. A prominent theory on the evolution of aging, which attributes special importance to genes that are advantagous in youth but are deleterious later on, is discussed in terms of regulatory genes that reduce DNA repair as cells differentiate to the postmitotic state. Finally, we hypothesize that the factors which determine maximum longevity of individuals in a population are the rate of occurrence of DNA damage, the rate of DNA repair, the degree of cellular redundancy, and the extent of exposure to stress.