Agent Orange was the most common herbicide used in the Second Indochina War in the course of military operations in the former South Vietnam. Agent Orange is contaminated by the carcinogen 2,3,7,8-tetrachlorodibenzo-para-dioxin (TCDD) in mean concentrations of 2 mg/kg. After much dispute of a causal association between exposure to herbicides containing TCDD and occurrence of soft-tissue sarcoma and non-Hodgkin lymphoma, two simultaneous case-control studies were set up in Vietnam to examine possible relationships. Subject recruitment is ongoing, with target numbers of 150 cases of soft-tissue sarcoma and 150 cases of non-Hodgkin lymphoma and diagnoses at the Cancer Center at Ho Chi Minh City, Vietnam. Two hospital controls are matched to each case. As in other studies of cancer in persons occupationally or otherwise exposed to herbicides and their contaminants, evaluation of past exposure of the recruited subjects is among the most complicated issues. Because accurate records are usually unavailable, surrogate measures of likely exposure are often calculated. As a first approach in our studies we used the Stellman and Stellman exposure index. The index is based on matching subjects' history of residence and the information on times and locations of Agent Orange spraying recorded on HERBS tape by the U.S. Army and taking into account the distance from the spraying as well as environmental and biologic half-life of TCDD. The exposure index is calculated in two centers, New York and Hanoi, with slightly different assumptions. In addition, samples of body tissues from the subjects (20 ml blood, 2 g adipose tissue, and tumor sections in paraffin blocks) are taken and stored. Their future analysis will provide additional source of exposure assessment. Strengths and weaknesses of both exposure measures are discussed in this paper.