About the tomb of the Marábut which we have mentioned above, there are frequent traces of building; and the tomb itself is constructed with the fragments of more ancient structures; while the beach and its neighbourhood are strewed with a quantity of pottery and glass. These ruins, although they now, with the exception of the Marábut and the forts, consist only of loose stones and imperfect ground-plans, appear to be more indicative of the site of an ancient town than those which we have mentioned at Guadigmata; and, if Graphara could be placed so near as twenty miles to Leptis Magna, they might probably be considered as its remains. The quadrangular forts which we have just mentioned as occupying the heights of Sidy Abdellàti, might in that case have belonged to a station attached to the town; and the port discovered by Captain Smyth at Abdellata (mentioned above) may then be taken as the one intended by Scylax. Without carrying the subject further, we may say, in conclusion, that Sidy Abdellàti has undoubtedly been a strong military station, whatever pretensions it may have to be considered as the site of Graphara. After leaving this place, the road led us, through the valley of Selîn, to a tolerably wide stream called Neggázi, which, winding between the hills, gave an unusual interest to the view. We continued our route for a short time along its banks, and then ascended the range of hills called Sélem, which branches off from the Terhoona[11] range and extends to the sea. From the top of Sélem there is an extensive view westward, over the plain of Jumarr, as far as the sandy desert; and on the eastern side of the ridge there is another view, equally imposing, over the plain of Lebida; so that in spite of the torrents of rain which still continued to deluge us, we could not help stopping occasionally to admire them.