Our course during the day lay between NW. and NNW., keeping for a short distance along the course of the stream. At two or three points we saw traces of rock dwellings, not nearly so well preserved as those of Aïn Tarsil. Before long the walls of rock subsided on each side of the stream, which bent in a westerly direction, while our track ascended gently amid undulating downs, till we reached a point commanding a wide unbroken view of the northern horizon. To the eye the country before us seemed almost a dead level, but there is a decided general slope from south to north, as we more fully ascertained in the evening, when we found that we had descended fully 1,500 feet during the day’s ride. The outline of the Iron Mountain (Djebel Hadid) was now clearly traceable in the distance, the highest part bearing about due NW. As we advanced there was a manifest improvement in the fertility of the soil, and for a space of five or six miles we rode amidst cultivated fields, apparently the most productive that we had seen in South Marocco. The same conditions necessarily affected the wild plants of the country, and made it difficult to secure specimens of manageable dimensions of several interesting species. A great _Daucus_ (_D. maximus_ of Desfontaines), probably a luxuriant form, or sub-species, of the common carrot, grew to a height of four or five feet, and the flowering umbels were often more than a foot in diameter. More ornamental was a splendid _Centaurea_, three or four feet high, with very large heads of deep orange flowers, often tinged with purple, which we took to be altogether new. Subsequent examination made it doubtful whether it should be separated from a very variable North African species, _C. incana_ of Desfontaines, though much larger in all its parts, and differing in the colour of the flowers. We halted for luncheon near a village called Hazarar Assa, standing, as we were assured, at the frontier, between the provinces of Mtouga and Shedma. In Beaudouin’s map those provinces are separated by an intervening strip of territory, apparently belonging to the Ouled Bou Sba tribe. As political divisions in this country are subject to frequent alteration, the map may have been correct at the time it was made. Maize was grown near the village, although no means for irrigation were apparent. As the growth of this plant is in general absolutely dependent on a frequent supply of water to the soil, we could only infer that we had come within the limits of the coast climate, and that as a general rule rain cannot there be very unfrequent during the months of April, May, and June. During the halt the villagers reported that fighting was still going on in the neighbourhood, and a few gunshots were heard in the distance; but, as our course led us away from the scene of action, the result, if any, was never known to us.