The use of chemical weapons in war violates the Geneva Protocol of 1925, but investigating alleged use of such weapons is difficult and a strict epidemiological approach has proved advisable. Two case studies in the use of this approach are presented. Testimony of those exposed is primary in such investigations and methods of inquiry must be designed so that consistency of responses can be used to establish validity of the testimony. Flaws in the methodology of the first investigation, conducted among the Hmong in Laos, and external difficulties affecting the outcome of the second investigation, conducted among the Kurds in Iraq, are detailed. Problems include leading questions, lack of time to conduct a proper survey, government opposition and the need for timely investigation., ABSTRACT: The use of chemical weapons in conflict represents a breach of international law as well as a grave violation of human rights. Investigating allegations of their use often is difficult. A basic tool is the survey interview. Experience has shown that a rigorous epidemiologic approach should be taken. A primary emphasis should be designing the study so that consistency of responses can be analyzed to judge the validity of the testimony. Only when the testimony can withstand this scrutiny is it possible to surmise the possible identity of agents employed. Securing samples of the putative agent is of obvious importance. Two recent investigations are discussed herein, one conducted by US Army medical researchers on allegations of chemical weapons use against the Hmong in Laos and another mounted by us on allegations of poison-gas attack against the Iraqi Kurds.