Passing into the Sahara from the Saharan Atlas, over the route which I followed, one encounters a great variety of topography, of which the most extensive may for the moment be classed as plains. The plains are divided into three well-marked regions, that of the dayas, the Chebka, and the Gantara (hamada). The topography is further diversified by oueds and their flood-plains and by low, flat-topped mountains. On the northern portion of the plains one encounters a sparse population of low perennials, and as Ghardaia is approached the plains vegetation becomes continuously poorer until at Ghardaia there appears to be none. On the hamada between Ghardaia and Ouargla areas are to be crossed, several kilometers in width, where perennials are wholly lacking. The decreasing plant population of the plains, until it entirely disappears, is entirely due to the increase in aridity as one goes from the mountains to and across the Ghardaia-Ouargla country. The low mountains are almost entirely barren. The flood-plains of the oueds, however, support a surprisingly luxuriant population of perennials. Should we contrast the topography and vegetation of the Algerian Sahara with that of southern Arizona we would find little that is similar and much that is different. The wide-stretching plains (bajada) of southern Arizona are well covered with perennials of good size. The water-courses are fringed with trees, and often an open forest is to be found on the flood-plains. The low mountains have a fairly dense plant population, partly of trees, and the lower mountain slopes are often covered with a mixed flora of shrubs and trees. It may be said that there is probably no large area in southern Arizona, where the soil conditions are favorable for plants, where the water conditions are too meager to support a perennial flora of some sort. The greater aridity of the northern portion of the Sahara is evident, therefore, from the great contrast in its flora. In crossing the plains of southern Algeria one is likely to call a region barren when close inspection will show that this is not the condition. In fact, it was found that areas on the plain, 16 by 16 meters in extent, carried as many as 330 or more perennials, although a casual glance did not reveal the presence of any conspicuous vegetation.[39] Both of these conditions are the immediate result of the small rainfall. The reason for the large number of plants in certain areas, as above noted, lies in their small size, since it would probably be difficult to find an equal number on this area were the plants as large, for example, as in southern Arizona. The fact that the perennials are inconspicuous is in part because they are small and in part because the leaves are either absent or greatly reduced.