New linguistic coinage can signify new practices and fresh perceptions in science: descriptors therefore are not trivial. Here, we consider the shifting valence of allergic and autoimmune in conceptions of experimental encephalomyelitis (EE). Ehrlich s dismissal of the relevance to disease of autoimmunity resulted in its long struggle for recognition notwithstanding the convincing attribution in 1904 of the hemolysis of paroxysmal cold hemoglobinuria. Yet allergy did take hold because of its assumption that harmful effects could be ascribed to an extrinsic agent against which immune responses were supposed to be directed, in line with contemporary microbiological research. In 1885 the history of EE began with Pasteur s anti-rabies vaccine, dried virus-infected rabbit spinal cord, with use occasionally inducing a post-vaccinal encephalomyelitis (PVE). From 1933 to 1935, PVE was investigated by Rivers who reported that some monkeys immunized with normal rabbit CNS extracts developed an inflammatory demyelinating EE and anti-brain antibodies: no cause was attributed. In the 1940s Freund developed an adjuvant that greatly potentiated immunization and in 1947 this was applied to animals immunized for EE: induction was accelerated and the disease was called E allergic E , initiating the EAE acronym. As recorded, the study of autoimmune disease leapt from nothing in 1945 to a vigorous field in the 1950s . Yet researchers sedulously retained allergic in the EAE acronym until the1980s, long after autoimmune had become available to them. Eventually practitioners for whom autoimmunity had meaning influenced the transition to E autoimmune E as the laboratory analogue of human autoimmune multiple sclerosis.