We could quote several other allusions to the ruins from Portuguese, Dutch, and English sources, copied one from the other, and all bearing the stamp of having come from the same fountain-head, namely, the Arabians, who told the Portuguese about them when they first arrived at Sofala. Our examination of the ruins confirms this in every respect. In our excavations we found Celadon pottery, Persian pottery, and Arabian glass, similar to the things found at Quiloa, where the Arabs also had a settlement. These objects represent the trading goods brought by the Arabians and exchanged with the inhabitants who lived in and around these ruins in the middle ages; but at the same time we found no trace whatsoever of the Portuguese, which would have been the case, as in other places occupied by them in Arabia and the Persian Gulf, had they ever been there. From these facts I think it is certain that we may remove from the Portuguese the honour claimed by them of being the modern discoverers of the ruins, an honour only claimed in the face of recent events, for De Barros is candid enough in telling us that his information came ‘from the Arabs who were there.’ Clearly to settle this question it is only necessary to quote a letter which I saw in the library at Lisbon, dated April 17, 1721, from the Governor of Goa, Antonio Rodrigue da Costa, to the king. East Africa was included then in the province of India, and the governor wrote as follows:— ‘(1) There is a report that in the interior of these countries many affirm there is in the court of the Monomatapa a tower or edifice of worked masonry which appears evidently not to be the work of black natives of the country, but of some powerful and political nations such as the Greeks, Romans, Persians, Egyptians, or Hebrews; and they say that this tower or edifice is called by the natives Simbabóe, and that in it is an inscription of unknown letters, and because there is much foundation for the belief that this land is Ophir, and that Solomon sent his fleets in company with the Phœnicians; and this opinion could be indubitably established if this inscription could be cleared up, and there is no one there who can read it. If it were in Greek, Persian, or Hebrew, it would be necessary to command that an impression be made in wax or some other material which retains letters or figures, commanding that the original inscription be well cleaned. ‘(2) At the same time it would be suitable to examine whether in that land is a range of mountains called Ofura, what distance it is from the coast or seaport, and whether it contains mines of gold or silver. ‘(3) In the same way it would be as well to inquire into the most notable names of those parts, mountains, chiefdoms, and rivers. ‘(4) To learn if the lands of Sofala are high or low, or marshy, or if they have any mountain ranges.’