In chemical carcinogenicity and drug-safety testing, a carcinogen is defined as an agent that when administered by an appropriate route causes an increased incidence of tumors in experimental animals as compared to unexposed control animals. Although a carcinogen may cause the appearance of tumors in organs where tumors do not usually occur in a given strain, the usual response is to increase the types of tumors seen spontaneously and to shorten the period of latency. The use of carcinogenesis experiments for research and safety assessment requires properly designed and well-conducted experiments and a knowledge of background data and variations in tumor incidences of control animals. Many factors can influence the reported incidences of spontaneous tumors. These include species, strain, sex, age, and source of the experimental test animal; study duration; extent of the pathology examination; dietary and environmental conditions; qualifications and experience of the study pathologist; diagnostic criteria and nomenclature conventions; and quality assurance and review procedures. This paper discusses several factors which may influence the incidence of tumors in control and test animals, and provides examples to illustrate the potential for these factors to affect the data.