g tree, with a thick, clumsy,
crooked, and deeply ringed stem. It was not a bit like either of the
palm-trees they had already cut down. Its trunk was not over ten or a
dozen feet high, but then, such leaves! They were not pinnated like
those already described, but what is termed "entire," that is, all in
one piece, and thirty feet in length by full five in width! Fancy two or
three dozen of these gigantic leaves standing up almost erect from the
top of the thick trunk, and you may form some idea of the "bussu" palm.
There are many palm-trees whose leaves are used for thatching houses,
but of all others for that purpose the bussu is the best.

These great fronds have a mid-rib, and from this, on both sides, run
veins in a diagonal direction to the edge. When they are used for thatch
the leaf is split up the mid-rib, and then each half is laid upon the
rafters, not straight, but in such a way that the veins of the leaf will
lie in a vertical direction, and thus serve as gutters to guide the
rain-water down the roof. A very few leaves will thatch a house, and a
covering of this kind, when properly laid on, will last for ten or
twelve years. So much are the bussu-leaves prized for thatch, that the
Indians, in parts where this palm does not grow, often make a canoe
voyage of a week to procure them!

The spathe which contains the flowers is also put to many uses. It is of
a long spindle shape, of fibrous, cloth-like texture, and brown colour.
The Indians use it as cloth. It makes an excellent bag, in which the
native carries his paints or other articles; and a large one, stretched
out, makes a very comfortable cap. Indeed, Guapo used the first spathe
he laid his hands upon for this very purpose.

There remained now to be found some palm-tree that would split easily,
and make laths for the roof, as well as planks for the door, shelves,
and benches. They soon discovered the very palm for these purposes. It
was one of the genus _Iriartea_, and known as the "pashiuba" palm. It
was a tree that differed from all the others in its aspect. It was a
noble-looking tree, rising, with a smooth stem, to the height of seventy
feet. At its top, there was a sheathing column swollen larger than the
stem, and not unlike the sheathing column of the catinga already
mentioned, except that that of the pashiuba was of a deep green colour.
Its leaves, however, differed materially from those of the catinga. It
is true, that, like them, they were p