Exposure to radiant heat caused dogs to pant and lose water by evaporation at rates of 40-70 g/hr. 2. When water was offered at intervals during the heating, the dogs drank at about half of the opportunities. The individual drinks were small but, by their repetition, loss of water by evaporation during heating was approximately matched by drinking. 3. Water given by stomach tube reduced drinking during a subsequent period of heating. 4. When water was offered more than 15 min after the end of a period of heating, after panting had ceased, drinking occurred only if the water loss exceeded 50-70 g, about 0-6% of the body water. This is regarded as drinking due to loss of water, beyond a threshold of dehydration necessary to stimulate drinking with the dog at rest. When water was offered during heating, drinking occurred with dehydration less than this threshold. 5. The drinking produced by heating was similar to that produced by running (O'Connor, 1975). When the animal ran under heat, panting was more severe and the water loss greater (85-150 g/hr); it was approximately matched by more drinking.
