The risk of developing a second primary cancer was studied among 171,749 men and 208,192 women who were reported to the Danish Cancer Registry between 1943 and 1980. Only those who survived at least 2 months were included in the analysis, and more than 1.7 million person-years of observation were accrued. Altogether, 15,084 second primary cancers developed in organs other than the initial cancer site [relative risk (RR) = 0.99]. Adjustment for possible underreporting of multiple primary cancers increased the RR to 1.06. The overall RR of a second cancer developing for all sites was 0.91, but interpretation of this risk is difficult because new tumors arising within the same organ are generally not recorded in Denmark. The RR for all sites increased with time from 0.94 during the first decade of follow-up (excluding the first year) to 1.13 among 30-year survivors. Patients below the age of 20 years when first diagnosed with cancer experienced significantly increased risk of developing a second cancer. Elevated risks were also observed for sites thought to have a common etiology. For example, cancers of smoking-related sites were increased in both directions for cancers of the oral cavity, respiratory tract, and urinary organs. For cancers suspected to have a hormone- or dietary fat-related association, significant reciprocal relationships were seen among cancers of the endometrium, ovary, and colon. Cancer treatment probably is an important factor in second cancer development, even when judged indirectly in the present study. For example, radiotherapy may have been responsible for an elevated risk of subsequent cancers of the thyroid, breast, colon, rectum, bladder, connective tissue, and hematopoietic system in long-term survivors. Chemotherapy may have increased the risk of subsequent leukemias. Our data further indicate that cancer patients have no general susceptibility to develop new malignant tumors, although high rates may be found for particular sites sharing common risk factors. Conversely, the occurrence of one cancer does not appear to protect against developing a new cancer.