To develop and assess effective instructional strategies and track conceptual understanding in large classes, it is convenient to use multiple choice diagnostic tests. Each diagnostic test item typically has one correct response with a series of incorrect responses representing common misconceptions. The Conceptual Inventory of Natural Selection (CINS) is a diagnostic test that assesses understanding of ten concepts related to natural selection: biotic potential, stable populations, limited natural resources, limited survival, variation within a population, variation inherited, differential survival, change in populations, origin of variation, and origin of species.  The identification of many common misconceptions, the development and field testing of items, and determination of their validity, readability, reliability and factor analyses will be described. We observed significant correlation between student performance on the posttest CINS and their end-of-semester interviews on the same topics. It appears that the multiple-choice test and the interview both reflect the students' underlying conceptual understandings.  Comparison of results for diverse student populations indicates that the most challenging topics are random origin of variation, origin of species, and change in a population over time.