This project compares various bibliometric measures and scientists' own judgments. Publication and cittion data are compiled for two cohorts of chemists awarded Sloan Fellowships. Citation patterns differ substantially between most cited papers and those these authors identify as their best. Theoretical, empirical, and methodological papers are contrasted as well. In addition, temporal citation patterns show that recognition spreads beyond the research area of a particular paper to yield cross-disciplinary citation surprisingly rapidly. Results suggest the utility of studying citation patterns among the Institute for Scientific Information Subject Categories, but also caution against equating publication and citation counts with scientific progress.