Ii6 EFFECTS OF MOISTURE AND HEAT. 



Africa, and the constant intercourse that has been going on 

 from time immemorial, it would be strange if there was not 

 some little connection between the languages of island and 

 continent, and perhaps a few roots commencing with the 

 consonants nrj and nj are from a different stock to the great 

 mass of the language. 



Then again, if all the different tribes of Madagascar are 

 from a common region, as one seems forced to conclude from 

 the unity of their language, how is it that such differences 

 exist in colour and physique, &c. ? Possibly, as seems sug- 

 gested by what has been already remarked with regard to the 

 Sih^naka, the colder climate of the highlands of the interior 

 has to some extent modified the Hovas as regards colour, 

 while moisture and heat combined have tended to darken the 

 other tribes living in the lower and warmer regions of the 

 country. But although this may explain some of the varia- 

 tions in colour and physical conditions, it cannot be con- 

 sidered adequate to explain the many other differences found 

 among the inhabitants of Madagascar. How are these to be 

 accounted for ? and where did they originate ? 



May it not be that we have in Madagascar, as in the Mala- 

 yan and Polynesian Archipelagoes, two races represented — 

 one, an olive or light-brown people closely connected with the 

 inhabitants of Eastern Polynesia (from the Sandwich Islands 

 through the Samoan, Marquesan, Society, Paumotu, and otlier 

 groups, down to New Zealand) ; and also a darker race, allied 

 to the Melanesian tribes inhabiting Western Polynesia, from 

 Fiji to New Guinea ? The mixed character of the words 

 comprised in the Malagasy vocabulary seems to favour this 

 suggested mixed origin. Taking Mr. Wallace's " List of One 

 Hundred and Seventeen Words in Thirty-three Languages 

 of the Malay Archipelago," we find that vdrona (bird) is from 

 the Malay and South Celebes languages ; dnaha (child), besides 

 being found in Malay, Javanese, and Celebes, is also used in 

 Ceram, Sanguir (Philippines), and by the Sea Gypsies ; oiio 

 (cocoa-nut) is found in Bouru, Amboyna, Ceram, and all 

 over the Polynesian groups inhabited by the lighter-coloured 

 races ; atdcly (egg) is used in Mysol, Gilolo, and Ceram ; and 

 so on. And an examination of Polynesian vocabularies would 



