Various concepts of thermoneutrality were considered for a proposed study of the role of hypothalamic amines in temperature regulation of rats. The classic definition, the ambient temperature over which metabolic rate is minimum and constant, gave a range of approximately 28 to 32'C. However, within this temperature range rats were inactive, the inactivity apparently representing a behavioural response to heat stress and itself responsible for the reduced metabolic rate; certain thermoregulatory effectors were also activated to increase heat loss. Therefore an alternative range, 180O± 1-9 (mean ± S.D.) to 28-1 ± 1 0°C, was defined in which rats displayed normal activity, behavioural thermoregulation being absent.For a study of the thermoregulatory effects of intrahypothalamic infusions of amines in conscious unrestrained rats, knowledge of their thermoneutral range is essential. Various concepts of thermoneutrality have been advanced. The classic definition, which considers only metabolic rate, is the ambient temperature range over which metabolic rate is minimum and constant. This gives a range of 28-33°C for rats housed at 23-25°C [Herrington, 1940;Swift and Forbes, 1939]. Outside this range, metabolic rate is elevated at the lower limit by enhanced heat production and at the upper limit by increased motor activity due to escape behaviour and thermoregulatory grooming (saliva spreading). However, this definition was considered unsuitable for the purposes of the proposed study, since within the temperature range defined (28-330C), body temperature is elevated above normal [Hainsworth, 1967; Hellstrom, 1975], the tail vessels are markedly dilated [Rand, Burton and Ing, 1965] and saliva spreading has been reported [Hainsworth, 1967[Hainsworth, , 1968; these effects would obviously distort the thermoregulatory responses to the amines. Clearly, a rat attempting to maintain a normal body temperature by increasing evaporative (saliva spreading) and non-evaporative (a sustained tail vasodilatation) heat loss is not in a 'neutral' condition [Mount, 1974]. A definition which also considers behavioural thermoregulation is likely to be more appropriate, for example that of Bianca [1974]: the range of ambient temperature in which the animal neither combats cold by raising its heat production nor heat by evaporative heat loss, and in which behavioural thermoregulation is normally absent. In an attempt to delineate this range in rats, metabolic rate and activities of thermoregulatory effectors (shivering, tail vasomotor tone and behavioural thermoregulation including saliva spreading) were monitored at different ambient temperatures.143