148 



H. B. GUPPY, M.B., ON POLYNESIAN PLANT-NAMES. 



plants are concerned, the Malagasy may be referred to Mela- 

 nesian stream of migration rather than to the Polynesian. 

 Nearly half of the names are to be found in the Philippine 

 Islands, and there is nothing in these data to connect the 

 Malagasy offshoot with any particular part of the Archi- 

 pelago, except, perhaps, with the Philippines. 



In the subjoined table I have given some rough results of 

 the numerical treatment of the subject. We notice there 

 that Malagasy contains the largest proportion of Indian 

 Archipelago plant-names and East Polynesian the smallest. 

 This appears at first sight not to be consistent with the 

 inference that Malagasy is more closely connected by its 

 plant-names with Fiji, or in other words with the Melanesian 

 migration, than it is with East and West Polynesian ; but it 

 must be remembered that whilst Malagasy was carried direct 

 to Madagascar in the course probably of a few weeks, the 

 Melanesian and Polynesian migration must have covered 

 many generations, and their peoples must have been exposed 

 to many vicissitudes on the way. That mere geographical 

 propinquity may count for little is shown in the case of the 

 Micronesian region whicli is less connected with the Philippine 

 languages by its plant-names than Malagasy. 



Indian 

 Archipelago. 



-nuT • Further 



Philippines. j^^j^_ 



Fiji 



West Polynesia 



East Polynesia 



Micronesia 



60 p.c. 

 55 „ 

 50 „ 

 60 „ 

 85 „ 



29 p.c. 

 37 „ 



28 „ 

 35 „ 

 46 „ 



11 p.c. 

 10 „ 



9 .. 



12 „ 

 19 ,. 



Note. — The Indian 

 Archipelago is here 

 regarded as includ- 

 ing the Philippine 

 Islands. 



