540 KEPOET— 1892. 



There are several points of interest connected witli this sacred en- 

 closure. The inner wall in front of the tower had been decorated with 

 courses of black slate ; a curious conduit about a foot square and regu- 

 larly constructed runs right through the thickness of the outer wall at 

 its thickest point. Similar and equally inexplicable conduits we found 

 about the temple on the fortress. Then there is the raised platform, ap- 

 proached by cement steps and a gateway just in front of the tower, 

 covered itself with a thick cement, into which a monolith had been 

 stuck ; this platform must have been for the king or officiating priest. 

 The whole of the sacred enclosure had been most carefully protected by 

 gates and buttresses : it is sunk a little lower than the rest of the build- 

 ing, and the outer wall is here at its strongest and thickest, and is, more- 

 over, decorated on the outside with the pattern which stops abruptly at 

 the place corresponding with the termination of the sacred enclosure in 

 the interior, and the summit of the wall for this portion only had been 

 decorated with large monoliths. 



The rest of the enclosure would appear to have been occupied by 

 private buildings also enclosed with circular walls, according, as Doughty 

 tells, to the ancient Arabian custom of combining their temples and 

 fortifications in one ; and El Masoudi also tells us that most of the 

 Sabsean temples were round. 



The monoliths form a marked feature : two over 13 feet in height 

 stand near the north-western entrance ; a third lay prostrate a few yards 

 away. The summit of the walls, too, was interesting ; it had been ap- 

 proached by steps from the pi'incipal entrance, was originally paved, and 

 hence formed a fine broad promenade with that curious decoration of 

 monoliths along it. 



Externally the walls are very fine, and the courses, from their extreme 

 evenness, give the building a neat and compact appearance. The pattern 

 which only decorates the part outside the long passage and the sacred 

 enclosure is in two rows let into the wall and formed by stones placed in 

 a chevron pattern, and must have required considerable skill to execute 

 as the building was proceeding. The circle of the walls is by no means 

 true, and the nature of the building is different ; as I have already men- 

 tioned, the portion from the main entrance and I'ound the sacred en- 

 closure is higher, thicker, and more carefully built; the rest was either 

 constructed at a later period or in a more careless manner ; it is impos- 

 sible now to walk on this portion of it, and it is rapidly falling into I'uins. 

 The middle entrance was evidently only a hole in the walls, or sally port, 

 probably of later construction, with wooden beams supporting the super- 

 incumbent structure, and which have entirely given way. 



Our excavations in this circular building were not attended with 

 success. It bore traces of Kaffir habitation up to within a recent date, 

 and they had evidently cleared out all traces of an earlier occupation ; in 

 some kitchen middens outside we came across traces of minor importance. 



The principal part of our work and our most interesting discoveries 

 took place on the hill fortress, to which we will now proceed, and the 

 labyrinthine nature of which can be best realised by a plan made by 

 Mr. R. W. Swan, and published for me by the Royal Geographical Society. 

 The hopje itself is of great natural strength, being protected on one side 

 by gigantic granite boulders, and on the south by a precipice from 70 

 to 90 feet in height, and on the only accessible side the inhabitants 

 constructed a wall of massive thickness like those of the ruins below : 



