62 . MAMMALS. 



some way, however faint, be connected -n-ith the other. That each has a common tj'pe to itself, WG 

 know to be the case. The Australian languages are all referable to one type. So are the African 

 (those north of the Sahara alwaj-s excepted). Mr. Burton* says of one of the tribes on the Cameroon 

 Mountains on the west coast of Africa : — " Their dialect is a branch of the great South African 

 family whose type is the Kaffir tongue." And Dr. Kirk informs me that the same thing is the 

 case with the languages of the Zambesians and those of the tribes stretching across Africa. The 

 Papuan language in like manner has many dialects. 



The inquiry leaves an alternative problem for the philologist to solve, viz., either to point out 

 the presence of some common elements of structure showing connexion between the languages of all 

 three countries, — ^Africa, Papua, and Australia, or else to show some discrepancy wholly inconsistent 

 with it. They have this basis to start from — that the foundation and structui'e of these languages 

 is different from that of the tribes north of tlie barriers above mentioned. 



Let us now see what peoples and countries may be referred to this black stock. Taking 

 Africa by itself, it is scarcely necessary to go into any argument to prove that all the tribes on that 

 continent south of the Saharan barrier belong to the same race. That may be safely assumed as 

 proved. Nearest to Afi-ica, and, only separated from it by a comparatively shallow connecting sub- 

 merged neck of land, lies Madagascar. The present state of its population requires to be sub- 

 jected to qualification before it can be admitted as relevant to this inquiry — a considerable portion 

 of it bearing strong evidence of colonization by the Malays. This is of comparatively recent date, 

 probably within, or not much beyond our own historic times, and of course cannot be taken into 

 account in speculating on the aboriginal population of the island. The result of this immigration 

 is not only a certain amoiint of Malay element among a portion of the Malagese, but the oc- 

 currence in their language of a considerable body of Malayan and Javanese words. That such 

 a colonization should have taken place is the more remarkable when we look at the great dis- 

 tance (3000 miles) from Madagascar to Java and the Malay peninsula. Ethnologists have attempted 

 to account for these peculiarities by supposing that a fleet of Malay pirates had been tempest- 

 driven on the coast of Madagascar, and, unable to find their way back, had at first been able 

 to protect and establish themselves, and afterwards becoming absorbed in the general population 

 by intermarriage, had, besides coimnunicating a portion of their blood, imparted some of their 

 knowledge, cultivation, and language. Mr. Crawford, speaking on this subject, f says : " The 

 people of Madagascar (that is, the aboriginal people) are not Malays, nor do they bear any resem- 

 blance to them. They are, in fact, negroes ; but negroes of a particular descrij)tion. They are 

 negroes in the same sense that Portuguese, and Laps, and Englishmen, Germans and Spaniard.s, 

 are European, and in no other." This is exactly what may be said not only of the Malagese, 

 but of all the other black tribes spread over the islands of the Southern Ocean. 



The Mauritius and Bourbon Islands may be dismissed as islands which were probably not 

 inhabited at the time of the original peopling of the other lands of which I speak. The Dodo 

 would never have survived to furnish even a solitary specimen or two to our museimis had the 

 islands on which it Uved been peopled by savages, whether black or white, or, I should rather say, 

 had they been peopled at all. 



Next, looking farther eastward, it will scarcely be disputed that, whatever objections there may 



* Burton, in " Proceedings of Geog. Soc." p. 241. (1862.) 



t Crawford, in " Troc. Geog. Soc." vol. ii. pp. 69, 70. (1862, 1863.) 



