Rats suffering from experimental allergic encephalomyelitis (EAE) were examined by a number of investigators whose assessments were compared. Considerable consensus of opinion was reached, especially when a standard, easily interpretable, scoring classification was used. More discrepancies occurred when a more expanded scale, differentiating between more clinical grades, was used. It is demonstrated that part of these discrepancies can be overcome by having the examinations done by the same investigator. Possible consequences of these findings for the assessment of clinical signs in EAE are discussed.