In this curiously decorated cave we found also many graves formed by plastering up holes in the rock with a hard kind of cement. We opened one of them, and found that the corpse had been wrapped in skins and placed here. In the centre of the cave is a large semicircular wall, entered in the middle by a rounded entrance; behind this is a sort of palisade of grass matting placed against poles, to protect it from the wind, and behind this are similar cement-covered graves. Now the present race do not bury in this way, but evidently come here at certain times to keep the place in order, and doubtless venerate the spot as the resting-place of remote ancestors. There are also several other graves on the flat space around Nyanger rock, piles of stones placed around a crescent-shaped wall, which is evidently a sort of rudimentary temple in which the sacrifices take place. On our march that day we passed several of these cemeteries in the open veldt far from any trace of habitation. They are generally placed on slightly rising ground, and always have the semicircular structure, which reminded us of the stones placed at the village of Lutzi, where the inhabitants collected to smoke and talk, protected from the wind. These spots are evidently still venerated, and form another of the many problems connected with the past in this district of Africa. I think they are the places to which Dos Santos alludes in the following paragraph, where he refers to the chiefs who ‘make grand obsequies to their predecessors, who are all buried there.’ In a memoir written by Signor Farao, governor of Senna in 1820, there is a curious testimony to this theory. He writes: ‘The mountains of Magonio (Makoni?), in Quiteve, were noted as the burial-places of the kings and queens of Quiteve, Gembe and Dombo. The remains were carried in procession to the caves, where they were deposited alongside the bones of former kings, and some of the most esteemed women of the deceased, or his secretary, and some of the great people, were sacrificed at the ceremony.’ Most of the granite kopjes in this district have been similarly fortified to Nyanger rock. Time would not permit of our visiting many of them, but I am certain that a careful investigation of this district would produce many valuable additions to the already large collections of Bushman drawings. The fortifications of these rocks are generally in rows of walls in terraces with narrow rounded entrances; they are all constructed in a rough manner, with irregular-shaped stones joined together with cement.