108 Sir Charles EJiot [May 19, 



hundred and thirty thousand square miles, contains a large African 

 element and several physical types of mankind. Yet authorities 

 agree in saying that a single Malay language, with no differences 

 greater than dialects, is spoken over the whole area, and it would 

 seem that this was originally the language of a relatively small body 

 of invaders. We may, therefore, conjecture that some strong 

 influence must have been at work in southern and central Africa, 

 which has succeeded in imposing such linguistic uniformity. 



If we could tell what that influence Avas, we should be in a fair 

 way to settle some of the most difficult questions of African ethno- 

 logy and history, but we have as yet little but theories to guide us. 

 But I think that in considering this mystery we should bear in mind 

 that South Africa contains another mystery which has lately attracted 

 considerable attention. This is, the existence in the country between 

 the Limpopo and the Zambesi of a considerable number of ancient 

 buildings, of which the mins called Great Zimbabwe are the best 

 known. The most recent authorities state that there are not less than 

 three hundred distinct groups of ruins in Rhodesia, and there are pro- 

 bably more in Portuguese territory. These iTiins are of different dates, 

 and much obscurity still involves the whole subject ; but it is quite 

 clear that there must have existed in this district during many 

 hundred years, and perhaps much longer, a forgotten civilisation far 

 superior to anything else that we know of in Africa south of the 

 Equator— a civilisation which was at first foreign, and afterwards, as. 

 it would seem, assimilated by some x\frican race. It appears to me 

 that there is a strong antecedent probaljility that there is a connection 

 between this civilisation and the diffusion of the Bantu languages, 

 and perhaps further researches in this quarter may throw some light 

 on their origin. The main objections to seeking for that origin near 

 the Zaml)esi are, that some authorities hold that the most archaic forms 

 of the Bantu languages are those spoken in Uganda and near lake 

 Victoria, and that many Bantu tribes have a tradition that they came 

 from the north. The linguistic evidence as to the relative antiquity 

 of these languages does not seem to me conclusive, and the most 

 decided traditions as to a northern origin seem to prevail among the 

 natives of our South African Colonies, and would be explained by a 

 southward migration from the country near the Zambesi such as must 

 have happened on any hypothesis. But I have no desire to dogmatise 

 on a matter which is still so obscure. It only seems to me that if in 

 one part of South Africa we have proof of the existence of a mys- 

 terious but long-continued ancient civilisation, it is probable that this 

 civilisation is connected with the exceptionally wide diffusion within 

 and all round the same area of a particular group of languages. 



The chief foreign influence which has affected East Africa in the 

 past has been that of the Arabs, or at least of the inhabitants of the 

 Arabian peninsula. Witli the exception of the Portuguese no Euro- 

 peans paid any attention to these regions until the latter part of the 



