With much acuteness of observation, William recognized the fact that the sun’s influence on the denser air of low altitudes is far more potent than it is on the rarer strata above.[735] Though heat comes from the sun, he said, it is not apparent until it becomes mingled with humidity. In valleys the air, lying stagnant and damp, is easily heated, whereas the dry upper levels remain cold even though the sun’s warmth passes through them. The presence of this coldness explains why snow is found on the summits of the highest mountains, for the belief that mountain snow is due to cold north winds William branded as false, observing that snow often occurs on the south as well as on the north sides of the peaks. Robert Grosseteste also held that the air at high altitudes is much colder than it is near the surface.[736] This, he said, was because the heating effect of the sun’s rays is inoperative on account of the transparency of the medium. At the surface heating takes place as a result of reflection and condensation of the solar rays.[737] The cold air at high levels explains the origin of perpetual snow on mountain tops. Hail is generated in these strata, rain at lower levels. Robert cited as proof of this the fact that birds of prey fly high in summer to cool off and that cranes and many other birds descend into the valleys to escape the icy chill but fly up the mountain sides to avoid the heat.[738] UPPER LEVELS OF THE ATMOSPHERE In contrast with these opinions of William of Conches and Robert Grosseteste, which were based apparently on more or less direct observation, we find echoes in our period of a doctrine that had its roots in classical mythology—the doctrine that above a certain height on mountain peaks the air is undisturbed by wind and unsullied by clouds.[739] Hermann the Dalmatian hints at this in his _Liber de essentiis_. In the course of a discussion of the dimensions of the habitable area of the earth’s surface that had probably been suggested by the reading of Arabic works he explains that the living offspring of the earth require for the maintenance of life a certain heavy, “greasy” terrestrial vapor which, “as Aristotle determined from the height of Olympus, does not rise more than sixteen stades above the earth’s surface. Here consequently would seem to be the upper limit of our habitable zone. Possibly this might be measured by means of the rainbow, which, according to the description of Hipparchus, reaches from the clouds themselves down to the surface of the earth. But since Hipparchus’ description is not accurate nor is the figure of the rainbow a semicircle, we leave the matter for whosoever may wish to prove it.”[740] Peter Alphonsi, who was also influenced by Moslem thought, placed the upper limit of the clouds at sixteen miles,[741] a figure which may have been derived from the same origin as Hermann’s sixteen stades. Peter Comestor inserted in his _Historia scholastica_ some