Even during its heyday, American behaviourist psychology was repeatedly criticized for the lack of diversity in its experimental subjects, with its almost exclusive focus on rats and pigeons. This paper revisits this debate by examining the rise and fall of a once promising alternative laboratory animal and model of intelligence, the raccoon. During the first two decades of the twentieth century, psychological investigations of the raccoon existed on the borderlands between laboratory experimentation, natural history and pet-keeping. Moreover, its chief advocate, Lawrence W. Cole, inhabited the institutional and geographic borderlands of the discipline. This liminality ultimately worked against the raccoon's selection as a standardized model during the behaviourist era. The question of raccoon intelligence was also a prominent topic in the contemporaneous debates over the place of sentiment in popular nature writing. Although Cole and others argued that the raccoon provided unique opportunities to study mental attributes such as curiosity and attention, others accused the animal's advocates of sentimentalism, anthropomorphism and nature faking. The paper examines the making and unmaking of this hybrid scientific culture as the lives of experimenters and animals became entangled.