44 
POLYNESIAN RESEARCHES. 
and the flower is contained in a sheath or spathe. 
There are several varieties; for thirty-three of 
which the natives have distinct names; and, as 
the plant is found to thrive best in moist situa¬ 
tions, it is cultivated in low marshy parts. A.large 
kind, called ape, arum costatum, which is fre¬ 
quently planted in the dry grounds, is also used in 
some seasons, but is considered inferior to the 
taro. 
All the varieties are so exceedingly acrid and 
pungent in their raw state, as to clause the greatest 
pain, if not excoriation, should they be applied to 
the tongue or palate. They are always baked in 
the same manner as bread-fruit is dressed; the 
rind, or skin, being first scraped off with a shell. 
The roots are solid, and generally of a mottled 
green or gray colour; and when baked, are pala¬ 
table, farinaceous, and nutritive, resembling the 
Irish potato as much as any other root in the 
islands. 
The different varieties of arum are propagated 
either by transplanting the small tubers, which they 
call pohiri , that grow round the principal root, or 
setting the top or crown of those roots used for 
food. When destitute of foreign supplies, we 
have attempted to make flour with both the bread¬ 
fruit and the taro, by employing the natives to 
scrape the root and fruit into a kind of pulpy 
paste, then drying it in the sun, and grinding it in 
a hand-mill. The taro in this state was sometimes 
rather improved, but the bread-fruit seldom is so 
good as when dressed immediately after it has 
been gathered. 
The uhi , or yam, dioscoria alata y a most valuable 
root, appears to be indigenous in most of the South 
Sea Islands, and grows remarkably well. Several 
