During the Mauritian period of Dutch Brazil (1637‑1644), a great deal of information about the biota of northeastern Brazil was obtained, consisting of both written records and paintings. Among them is an eel-like fish, depicted in two paintings labeled “Caramurû” and “Caramuru”, whose taxonomic identity is controversial. One of them, attributed to Albert Eckhout, is part of the Theatrum Rerum Naturalium Brasiliae collection, stored at the Jagiellonian University Library, in Poland. This painting was possibly a model for the second “Caramuru”, of unknown authorship, which is deposited at the Archive of the Russian Academy of Sciences. These paintings are doubtfully identified as the snake eel Echiophis intertinctus (Ophichthidae, Anguilliformes), a proposal likely induced by the vulgar name “caramuru”, which is applied to muraenids and ophichthids in Brazil. After careful examination of these two paintings regarding the anatomical details depicted, we concluded that the fish corresponds, with great certainty, to Gobioides broussonnetii (Gobiidae, Gobiiformes), popularly known as “aimoré” and “tajasica”. Furthermore, we suggest the possibility that the written counterpart of these paintings is the description associated with the woodcut of the fish labeled as “Tajasica” in the Historia Naturalis Brasiliae (Marggraf in de Laet, 1648). The confusion probably stems from mismatches between Georg Marggraf’s descriptions and the images of organisms produced at the time, perhaps before the return of Johan Maurits van Nassau-Siegen’s entourage to Europe. In an attempt to support our conclusions about the identity of that fish, as well as for historical purposes, we also discuss the vernacular names applied to G. broussonnetii and the authorship of the notes made on those paintings.