This work aims to offer a new methodological approach to intellectual exchange in the medieval Mediterranean. In the absence of abundant textual evidence, this work explores the transmission of scientific knowledge from Greek and Arabic into Latin during the eleventh and twelfth centuries through direct textual translation as well as unwritten, typically material, exchanges. It approaches the so-called twelfth-century translation movement of Greek and Arabic science into Latin in three parts, which each touch on a different branch of medieval science that was transmitted into Latin through medieval Sicily. The first part examines paratextual diagrams in medieval manuscripts of the Classical work on mathematics and astronomy, Ptolemy’s Almagest. Working across Latin, Arabic, and Greek, it traces a different route for the transmission of the mathematical diagrams than for the translation of the text itself. In the second part, it moves away from direct translation, turning to the production of Latin pharmaceutical manuals that lack a direct antecedent in another language. For one of these, Circa instans of Matthaeus Platearius, it first considers on both a holistic and granular scale how the Latin text drew on influences in Greek and Arabic that would have been available in Sicily in the twelfth century. This comparison suggests that some information about pharmacology was transmitted orally or experientially. The next section compares the individual substances included in Circa Instans to Latin and Judaeo-Arabic trade records for Sicily during that period, in order to determine whether and how information about these goods as medicines could have moved through trade, which otherwise considered them to be supplemental materials for the textile industry. T he third part is focused on copper, one of the materials mentioned in both pharmaceutical manuals and trade records, which also has a significant presence in extant objects from Norman Sicily. By systematically surveying these extant objects, as well as the treatment of copper in alchemical manuals from the period, this work considers the different information that was conveyed through the material presentation of copper than through its treatment in alchemical treatises. In these analyses, this work demonstrates that a study of medieval science can benefit from: considering a broad range of sources, both in language and medium; navigating carefully through assertions of what the knowledge being transmitted constituted; and reevaluating assumptions about the role that textual translation played in transmitting knowledge of science.