After several years of trial and error, I have developed a method to generate consistently good library research papers in undergraduate science classes, papers that show thought and integrate information. It is based, in part, on treating the student papers as journal article submissions. There are five steps to the procedure which are spaced throughout the semester to ensure that the papers will not be done at the last minute. The steps are: (1) selecting a topic; (2) computerized on‐line library reference search; (3) first draft of paper; (4) paper review by a classmate; and (5) submittal of revised paper. Each step requires a written document and has a class‐discussion component. A side benefit of the discussion component is that it helps to set students at ease about participating in class discussions, which carries over to discussions about the scientific content of the class. The resulting well‐written papers are usually a pleasure to read and grade.