If asked to define fever, most physicians would offer a thermal definition, such as "fever is a temperature greater than...." In offering their definition, many would ignore the importance of the anatomic site at which temperature measurements are taken, as well as the diurnal oscillations that characterize body temperature. If queried about the history of clinical thermometry, few physicians could identify the source or explain the pertinacity of the belief that 98.6 degrees F (37.0 degrees C) has special meaning vis-à-vis normal body temperature. Fewer still could cite the origin of the thermometer or trace the evolution of modern concepts of clinical thermometry. Although many would have some knowledge of the fundamentals of thermoregulation and the role played by exogenous and endogenous pyrogens in the induction of fever, few would have more than a superficial knowledge of the broad biological activities of pyrogenic cytokines or know of the existence of an equally complex and important system of endogenous cryogens. A distinct minority would appreciate the obvious paradoxes inherent in an enlarging body of data concerned with the question of fever's adaptive value. The present review considers many of these issues in the light of current data.