ORIGIN OF RIVERS To return from this digression to the vexed question of the origin of super-terrestrial rivers, we find that the circulation of water from the sea either by underground passages or by rain was not regarded by the majority of ancient thinkers as sufficient to account for the huge volumes of water that rivers constantly pour into the sea. Plato had thought that there were enormous reservoirs in the interior of the earth which served to keep the rivers supplied,[129] but Aristotle rejected this hypothesis.[130] A reservoir as large as the entire earth, he said, would be necessary for the purpose. His explanation was worked out of the theory that one element actually may be transformed into another. In a relatively unscientific age what is more natural than to believe, when one sees soluble substances passing into solution in water, that they actually become water? Or when one sees the condensation of invisible vapor into clouds and of clouds into rain, that the air is actually turning to water? Aristotle, followed by Seneca,[131] argued that the air which penetrates into the internal cavities and recesses of the earth is chilled and liquefied by the cold encountered there, just as air seems to be condensed by cold in the outer atmosphere. Aristotle cited as a proof of this the supposed fact that most great rivers have their sources in mountains.[132] Mountains were to be looked upon as enormous elevated sponges exuding water on all sides. Aristotle concluded likewise that the northern part of the earth must be high and mountainous,[133] because many great rivers originate there. But, if the air is transmutable into water, why, then, was it not perfectly logical to suppose that the earth could also undergo a similar change? This as a theory to explain the origin of some of the water of rivers was clearly expressed by Seneca and, among the early Church Fathers, by Gregory of Nyssa.[134] The faulty character of Seneca’s scientific thought is seen in his failure to account satisfactorily for the logical demands of his theory, i. e. for the replacement of the land lost by its liquefaction. THE NILE FLOOD