IX 
AROUND SAO GABRIEL 
five times. It is the largest we met, being 4 
or 5 yards wide in its upper part, the depth now 
rarely more than to the knees, but in the flood 
averaging about 4 feet. It is called Ui'wa-igarape 
(or the river Arrow), and does not run directly into 
the Rio Negro, but into the Curicuriarf, proving 
the latter to deviate much in its upper part from 
its direction near the mouth. Hence also the 
Serra do Gama may be considered a continuation 
of the Serras do Curicurian', though there is appar- 
ently a great gap between them. The Uiwa has 
a sandy bottom and clear (not black) water. On 
its banks we chose a place to pitch our tents, 
having arrived as near to the serra as we judged 
convenient. In its sands and on rocks standing 
out of it, I got some interesting Ferns. Close 
by our resting-place (which we reached at i p.m.) 
was a large Loureira, at least 100 feet high and 
very straight. This we tapped and a good draught 
from it twice a day was my allowance whilst we 
stayed. The milk was thinner than I had before 
met with it, and the Indians say that the milk of 
all milky trees is more copious and flows more freely 
in the wet season than in the dry. There were 
also some very tall Assai's and Paxiiiba barriguda 
palms, perhaps over 100 feet high. 
My men (three of them) set to work to erect a 
couple of huts, one thatched with Assaf, the other 
with Paxiiiba. For each two trees were selected 
at a convenient distance for hanging the hammocks 
below, and to support the roof short sticks were 
tied across the trees, forming triangles. The huts 
were but just finished when the rain, which had 
been growling for some time in the distance, 
