the stream, one 20 m., the other 70 m. long. There a _corrideira_
occurred soon after we had negotiated a dangerous rapid--dangerous
because of the number of intricate rocks which forced the canoe to
describe a snake-like dance like a double S, bumping and swerving with
such force from the restless waters underneath, that it was all we could
do to prevent her turning over.

In a basin 700 m. wide which was further crossed, we admired a
picturesque rocky island of a beautiful emerald green colour in the
centre of the stream. An immense barrier of rock was on the north-east
side of this basin. Before we halted, absolutely worn out by the heavy
work of the day, we descended another troublesome rapid--fortunately that
time with no mishaps of any kind.

At five o'clock we made our camp in the only spot we could find that was
suitable; but no sooner had we landed than we were fiercely attacked by
millions of _sauba_ or _carregadores_ ants which gave us a lively time
during the entire night. Those ants, which were there absolutely in
millions, were from 1 in. to 11/4 in. in length, and possessed powerful
clippers on the head with which they bit us, giving intense pain. When
you had thousands of them climbing up your legs and over your body, and
dropping upon you from the tree branches which were alive with them, and
clinging to you with all their might once they had got you with their
clippers, you began to think what a fool you had been to leave your happy
home in England.

As I shall have an opportunity of speaking at greater length of the
_saubas_ later in this volume, I shall leave them now, merely mentioning
that during the entire night we were unable to sleep owing to those
brutes. And that was not all: we had many of our clothes, shoes, and
other articles entirely destroyed by them.

We called that place Camp Carregador. The nights had become by then quite
stifling and damp, the minimum temperature on July 21st being 63 deg. F.

No sooner had we started on our journey that day than we came to rapids.
A lot of rocks stood everywhere in the stream. The river after that
flowed in a snake-like fashion for 5,000 m. in a general direction
N.N.E., and was there comparatively free from serious obstacles. We came
to a triangular island 700 m. long--Ada Island--separated from a second
island by a channel 50 m. wide. This second island--Hugo Island--formed
an isosceles triangle of 800 m. each side. These two islands were
ev