The evidence favoring a role for prostaglandin E (PGE) as the neural mediator of the febrile response is reviewed and considered under five different essential criteria which would need to be satisfied, if such a role is to be accepted. These criteria are: the ability of intracerebrally microinjected exogenous PGE to cause fever; the detection of increased levels of endogenous PGE in the brain during the normal production of fever; the ability of substances that inhibit the production and release of PGE to block normal fevers; the ability of substances that are specific PGE antagonists to inhibit normal fevers; and the identification of a specific site and cell type for the release of PGE in response to the action of pyrogens. Evidence from the literature that supports these criteria is reviewed and presented in this format, and the conclusion is drawn that the evidence available is more than sufficient to support the initial hypothesis.