146 H. B. GUPPY, M.B., ON 



a great variety of sources during several years.* It does not 

 aim at completeness in any one respect, nor can it be claimed 

 to be always free from error. It presents, however, a more 

 extended view of the subject than, I think, has ever been 

 presented before. And yet to draw up a mere list of names 

 without further comment, and without any critical observa- 

 tions on their nature, is not a very satisfactory method of 

 dealing with the subject. The names, for instance, of the 

 sweet-potato only awaken curiosity, and it is obvious that in 

 order to learn the true meaning of the term " kumara," we 

 must discuss the history of this plant over the whole Pacific, 

 and extend our inquiries to the Indian Archipelago, and even 

 to Mexico and Peru, or even to India, if we follow Tregear, 

 though I think in that respect he is wrong. We must indi- 

 cate the localities where it is mostly used. We must look 

 into the antiquity of its culture, and we must be sure of its 

 uniform specific identity. De Candolle has pointed out the 

 uncertainty that surrounds its origin. From a large number 

 of facts I have formed the opinion that the sweet-potato is 

 like the yam, a very prosaic vegetable ; that it neither bears 

 the name of an Aryan deity, nor find its home in the heights 

 of Ecuador; that it has been for ages cultivated in the 

 Pacific, and has found its way there from the Archipelago in 

 a matter-of-fact fashion, in the sober society of the taro. 

 The history of the sweet-potato in the Pacific is the history 

 of its name " kumara," and a great deal more ; and the same 

 may be said of many other of the plants included in the 

 table. The discussion of the various questions connected 

 with the origin of the bananas in the Pacific must precede 

 any inquiry into their names, and the demarcation of the 

 great " Vundi " region is itself a task of no small labour.f 

 In order to deal at all satisfactorily with plants like Morinda 

 citrifolia and the turmeric we have to open up the question 

 of colour-names, to inquire into the history of dyeing pro- 

 cesses amongst savage peoples, and in the case of the tur- 

 meric to mark out the range in time and space of an ancient 

 ceremonial custom whilst we are investigating the distribu- 

 tion of its names. 



One curious result, which is but slightly illustrated in the 



* For the ludian Archipelago I have made free use of Filet's PlaiU- 

 hundig Woordenboek voor Nederlands Indie, Blanco's Flora de Filipinas, 

 Kern's Stamland paper before quoted, etc., etc. 



t This was first attempted in Codrington's Melanesian Languages. 



