THE POLYNESIANS AND THEIR PLANT-NAMES. 147 



table, is the frequent crossing of the tuber or root names. 

 This carries us back to that distant epoch when plants with 

 edible roots or tubers were known under a very few general 

 terms to primitive man over a v/ide region of the tropics. 

 The word " taro " in the form " tara" is applied by the Tas- 

 manians to Fteris esculenta, and according to ^Mitchell "tao " is 

 the name of a cichoraceous plant the root of which was 

 commonly eaten about the Darling. With the Maoris " roi " 

 is the general word for edible fern roots. In Cochin China 

 " ray " is a generic word applied to the edible Colocasice and 

 Alocasice. Then again " maraagu " or "• mamaku," the name 

 of an edible fern in New Zealand is a name of Tacca pinnati- 

 fida in the Solomon Islands, and apparently also in the 

 Marshall and Caroline groups. In the filial ects of British 

 New Guinea Ave find the connecting link between the names 

 of the sago palm in the Indian Archipelago and those of 

 lacca pinnatijida in Fiji and Polynesia, both farina-yielding 

 plants. There is also a singular group of tuber or root 

 names typified in the words " koko," " kuka," " kuku," 

 " kakau," etc., Avhich are commonly applied to plants of this 

 description in different localties. Thus " koko " in the 

 Solomon Islands is a word for the taro. In the Loyalties it 

 is given to the yam. Kuka is the Maori word for the edible 

 roots of the bulrush and of the Cordyliue. "Kuku" is the 

 general term for ferns in Fiji. '• Kaukau " is the Tongan 

 word for a kind of yam, and " Kakau" is the ancient name 

 of the Maories for the sweet potato. " Kokau " is a Pelew 

 word for the taro. " Lau'o'o " in Samoan is a species of Alo- 

 casia, and " 'Au'o'a " a yam. It is difficult to believe that 

 we have not here some primitive word for a wild edible 

 tuber or fern-root, although the direct linguistic connection, 

 as far as the vowel-changes are concerned, is often now not 

 possible. We will no doubt be able in time, from the 

 evidence of language alone, to frame the dietary of primeval 

 man in this part of the world. 



Before concluding this paper I would point out that by 

 giving a numerical value to the data in the table we can 

 obtain some interesting indications of the affinities of the 

 plant-names of each region taken collectively. I will first 

 take the Malagasy words. Out of 26 that admit of comparison. 

 22 occur in the Indian Archipelago, 15 in the West Pacific, 

 15 in Fiji, 12 in West Polynesia, 8 in East Polynesia, and 5 

 in Further India. It would appear from these figures that 

 as far as the names of several of the cultivated and littoral 



