Argument A growing cross-disciplinary literature has acknowledged the importance of verbal-visual interaction in the creation and communication of scientific texts. I contend that the proper understanding of these texts must flow from a hermeneutic model that takes verbal-visual interaction seriously, one that is firmly grounded in cognitive constraints and affordances. The model I propose has two modules, one for perception, derived from Gestalt psychology, the other for cognition, derived from Peirce's semiotics. I apply this model to an important but largely neglected text in the history of nineteenth-century science, Charles Lyell's The Geological Evidences of the Antiquity of Man, and to two accounts by well-respected historians of science, both of the same key discovery in quantum physics. In doing so, I hope to demonstrate the advisability of incorporating the exegetical practices my model entails into the everyday practices of historians of science.