Water is so much a part of life that it is impossible to imagine a terrestrial organism not provided with mechanisms to ensure sufficient supplies. Among the truly terrestrial vertebrates, the mammals, birds and reptiles have well defined, albeit incompletely worked out, mechanisms of thirst. The amphibia in general do not drink but they absorb water according to need across the skin. It is not clear whether they have a “water drive” activated by dehydration which would be analogous to thirst in the higher vertebrates, but despite mainly negative reports it seems probable that some amphibians, at least, do seek out water when they are dehydrated. It is interesting that prolactin, which has the effect of inducing immature newts to seek water and assume their reproductive form, has been reported to cause thirst and salt craving in human volunteers [1] and retention of sodium and water without compensatory reduction of either saline or water intake in rabbits [2]. The prolactin-induced “water drive” effect in the newt may have little to do with water balance since its purpose is almost certainly mainly reproductive. Nevertheless, such behavior, the burrowing that amphibia inhabiting the arid regions engage in in order to reach the damp subsoil, as well as the numerous reports that amphibia seek water, suggest that amphibia may experience a sensation akin to thirst in the higher vertebrates when dehydrated [3].Even among the fish there appear to be situations where drinking of water is provoked by need. In sea water the eel, a catadromous teleost, like other marine teleosts, drinks continuously in order to replace the water lost by osmosis across the body surface, but as soon as it enters fresh water it stops drinking. However, in fresh water it can be made to drink by infusing hypertonic saline (a cellular stimulus to thirst) or by removing blood (an extracellular stimulus to thirst) [4]. The eel therefore resembles the mammal in showing drinking behavior in response to both cellular and extracellular dehydration. Unlike the mammal, drinking by the eel appears to be reflex since it continues after removal of the prosencephalon and mesencephalon. There is little purpose in discussing whether the fish experiences thirst, a conscious sensation, or whether the act of drinking is entirely unconscious. Since the fish lives in water, the neural mechanisms needed to ensure that enough is drunk could be simpler than in a terrestrial animal where the much more complicated behavior of first seeking and then ingesting water is required. It is this complexity of behavior that could account for the encephalization of drinking in the terrestrial animal. What is important is that the fishes have already evolved the neural organization and other mechanisms necessary to induce intake of water in states of dehydration.The progression during phylogeny from essentially reflex drinking in the fish to the much more complicated behavior of seeking and consuming water shown by the terrestrial vertebrates is reflected in the ontogeny of drinking behavior [5]. Six-day-old suckling rats made thirsty by cellular or extracellular dehydration will readily swallow water placed in the mouth, but they are not capable of seeking water for themselves. They therefore resemble freshwater fish. When they are two weeks older, however, they show the same water-seeking behavior as thirsty adults. The development of adult drinking behavior does not appear to depend on individual postnatal experience though it is conceivable that quantitative differences between adults may stem from different postnatal experiences to deficits.