ON THE RUINS OF MASHONALAND. 543 



very lai'ge one, whicli is almost intact, and has a very finely chiselled sur- 

 face and well-rounded edges. There is one fragment on which I take it 

 some letters are rudely sci-atched, resembling somewhat the Ogham 

 characters or strokes on either side of a line to give them value. I have 

 not yet found anyone who can suggest any solution of this problem, and 

 I should have felt inclined to consider it hopeless had not Mr. Anderson 

 shown me a sketch of a stone fi'om Bechuanaland, taken thirty years ago, 

 whicli has characters on it closely corresponding to these. 



Many fragments of pottery of excellent glaze and workmanship were 

 also found ; the patterns on these are mostly geometric, but executed with 

 absolute correctness with a stamp, and there appear to be wheel-marks at 

 the back. Of the numerous implements of war we found many are 

 doubtless Kaffir and of a later date. One, however, is' an assegai with a 

 heavy plating of gold on it, which must belong to an earlier period, and 

 the spear-head of copper, heavily barbed, corresponds to weapons now 

 found amongst natives not at a great distance from this spot. The iron 

 bells, too, which we found in the ruins occur now on the Congo, hundreds 

 of miles away ; but the uniformity of design all over Africa and the con- 

 servative nature of these designs make it difficult to attach any date to 

 them. Everywhere in the ruins we turned up circular whorls of pottery, 

 some decorated and some plain, which at first we took to belong to 

 spindles for cotton, but some are too small and could never have been 

 used for that purpose. The circular soapstone object with knobs thereon, 

 which was brought, together with the bird shown to the Section, 

 by a Dutchman from Zimbabwe two years ago, Mr. Trimen, of the 

 Cape Museum, has kindly lent for exhibition in England. The first 

 impression it gives is that of a quern, but soft stone like this can never 

 have been used for the purposes of grinding, and all objects of soapstone 

 found in the ruins would appear to be for decorative purposes only. 



In one corner of this same part of the fortress, near the surface, we 

 found several fragments of Persian and Celadon china, doubtless used as 

 barter goods by the Arab traders who came up from the coast ; but no- 

 where during all our excavations could we come across any signs of coins 

 or traces of a cemetery for the burial of the dead. 



Perhaps the most interesting of our finds in this portion of the ruins 

 were those in connection with the manufacture of gold. Close under- 

 neath the temple stood a gold-smelting furnace, made of very hard 

 cement, with a chimney of the same material, and very neatly bevelled 

 edges, portions of which are now on their way home. Hard by, in a 

 chasm between two boulders, lay all the rejected quartz casings from 

 which the gold-bearing quartz had been extracted by exposing them to 

 heat prior to the crushing, proving beyond a doubt that these ruins, 

 though themselves far removed from any gold reef, were the capital of a 

 gold-prnducing people, who had chosen this hill fortress with its granite 

 boulders owing to its peculiar advantages for strategic purposes. 



Near the furnace, too, we found many little crucibles of a composition 

 of clay, which had been used for smelting the gold, and in nearly all of 

 them exist small specks of gold adhering to the glaze formed by the heat 

 of the process. There are tools also amongst our finds for extracting 

 gold, burnishers, crushers, &c., and an ingot mould of soapstone of a 

 curious form, which is still in use amongst the natives much fui'ther 

 north for ingots of iron. 



An interesting parallel to the ancient gold workings in Mashonaland 

 is to be found by studying the accounts of the ancient gold workings of 



