The fresh sheet of snow which had fallen on the Aït Mesan range within the last few days led us to what seemed an explanation of the inconsistent accounts given as well by travellers as by natives as to the existence of perpetual snow on this part of the Great Atlas. From its position between the Great Desert to the south, the Atlantic Ocean to the west, and the low country to the north, it is obvious that a range of mountains from 11,000 to 13,000 feet in height must frequently be the seat of violent atmospheric disturbances. Whenever these draw from the ocean currents of heated air, nearly saturated with moisture, into the upper region, the cooling effect consequent on rapid expansion must produce copious precipitation, and it is most probable that on the higher part of the range this, even in the hottest season, takes the form of snow. But, as we had seen, the snow melts with extreme rapidity under the almost vertical sun during the summer months; and hence one traveller may have seen the range thickly snowed even in the hottest season, while another, with equal truth, may describe it as almost completely bare. The state of things is such that a very moderate change in the physical conditions might easily lead to the accumulation of an annual surplus of unmelted snow, which is the first condition for the formation of glaciers. A mere increase in the amount of precipitation, with little change in the general conditions of temperature of this region, might produce glaciers reaching as low down as that whose moraine we saw at the head of the Aït Mesan valley.