102 OTAHITE CHAP. V 
seven feet high, and too bulky in proportion to its height ; 
the whole was neatly covered with feathers—white to re- 
present skin, and black to represent hair, and tallow on the 
head, where were three protuberances which we should have 
called horns, but the Indians called them tata ete (little 
men). The image was called by them Manne. They said 
it was the only one of the kind in Otahite, and readily 
attempted to explain its use, but their language was totally 
unintelligible, and seemed to refer to some customs to which 
we are perfect strangers. Several miles farther on we went 
ashore again, though we saw nothing remarkable but a 
burying-ground, whose pavement was unusually neat. It 
was ornamented by a pyramid about five feet high, covered 
entirely with the fruits of Pandanus odorus and Cratcva 
gynandra. In the middle, near the pyramid, was a small 
image of stone very roughly worked, the first instance of 
carving in stone that I have seen among these people. This 
they seemed to value, as it was protected from the weather 
by a kind of shed built purposely over it. Near it were 
three human skulls, laid in order, very white and clean, and 
quite perfect. 
We afterwards took a walk towards a point on which 
we had from afar observed trees of etoa (Casuarina equiseti- 
folia), from whence we judged that there would be some marat 
in the neighbourhood; nor were we disappointed, for we 
had no sooner arrived there than we were struck with the 
sight of a most enormous pile, certainly the masterpiece of 
Indian architecture in this island, and so all the inhabitants 
allowed. Its size and workmanship almost exceed belief. 
Its form was similar to that of marais in general, resembling 
the roof of a house, not smooth at the sides, but formed into 
eleven steps, each of these four feet in height, making in 
all 44 feet; its length was 267 feet, its breadth 71 feet. 
Every one of these steps was formed of white coral stones, 
most of them neatly squared and polished; the rest were 
round pebbles, but these, from their uniformity of size and 
roundness, seemed to have been worked. Some of the coral 
stones were very large, one I measured was 34 by 24 feet. 
