Technical limitations have precluded the continuous measurement of the deep body temperature of animals when unrestrained in an environment in which they occur naturally, or are traditionally husbanded. Existing knowledge of the nychthemeral (24 hourly) and seasonal variations in body temperature derives from discontinuous measurements of rectal temperature under field or laboratory conditions which have required varying degrees of restraint and handling. In consequence, despite innumerable laboratory studies of the physiological processes upon which mammalian homoeothermy depends, there is little reliable information on the degrees of thermostability which different species of mammals actually achieve under natural conditions.In the context of previous observations on the thermoregulatory physiology of sheep (Bligh, 1959(Bligh, , 1961(Bligh, , 1963 (Minett & Sen, 1945;Symington, 1960;Eyal, 1963) have been subject to the experimental limitations mentioned above.In principle, two techniques can be used to avoid the frequent attentions of the observer. An indwelling thermosensitive element can be coupled either to a miniature recorder attached to the animal, or to a portable radio transmitter. The miniature recorder has the advantage that it is not subject to the legal and technical difficulties which may limit the range of radio transmission. It has, however, the disadvantage that the record cannot be monitored. In fact, a satisfactory portable recorder of the required dimensions was not available when this study was planned.A suitable radiotelemetric system for the transmission of thermal information with an accuracy of within 0.10 C continuously for more than 100 hr over a range of 110 m was devised for this purpose (Bligh & Robinson, 1963;Robinson, 1964). This has permitted the collection of a