An analytical scheme that differentiates among the various types of cognitive and social functions of citations is used as the basis for an analysis of the results of a questionnaire designed to probe the citing behavior of a group of scientists who had cited one of four papers originating from a single biochemical laboratory. Even when papers fall within a relatively well-defined research area and are based on research conducted within a single lab, groups of scientists to which a given paper is of relevance can have quite distinct internal structures. We argue that the scientists'subjective reasons for citing a given work do not correspond with the actual argumentative uses of cited articles in citing texts. Scientists appear to regard the papers they cite as having a rather generalized significance even if in their citations they refer to very specific claims made in the original texts. The consequences of these findings for a theory of citation and for the use of citations in science studies are discussed.