To prepare students for participation in a pluralistic, democratic society, history curriculum should help them develop mature ideas about why multiple accounts of the same events exist. But how can we know if we are successful? In this article, we describe work on the design, validation, and piloting of a paper-and-pencil instrument called the Historical Account Differences (HAD) survey. It is intended to add to the tool set that researchers and practitioners can use to evaluate students' developing metahistorical conceptions. Data are presented from a pilot involving sixty-five eleventh-grade students. One group completed the survey once, whereas the other completed it prior to and after a curriculum unit designed to enhance their conceptions about multiple accounts of past events. Initially, the two groups' responses were not detectably different. However, students completing the two-week curriculum unit showed an increase in mature conceptions. The limitations of this work and plans for future research are discussed.