plains. I found that the mules usually arrived so late in the
afternoon or evening that I could not depend upon being able to write
at that time. Of course, if we made a very early start I could not
write at all. At night there were no mosquitoes. In the daytime gnats
and sand-flies and horse-flies sometimes bothered us a little, but not
much. Small stingless bees lit on us in numbers and crawled over the
skin, making a slight tickling; but we did not mind them until they
became very numerous. There was a good deal of rain, but not enough to
cause any serious annoyance.

Colonel Rondon and Lieutenant Lyra held many discussions as to whither
the Rio da Duvida flowed, and where its mouth might be. Its
provisional name--"River of Doubt"--was given it precisely because of
this ignorance concerning it; an ignorance which it was one of the
purposes of our trip to dispel. It might go into the Gy-Parana, in
which case its course must be very short; it might flow into the
Madeira low down, in which case its course would be very long; or,
which was unlikely, it might flow into the Tapajos. There was another
river, of which Colonel Rondon had come across the head-waters, whose
course was equally doubtful, although in its case there was rather
more probability of its flowing into the Juruena, by which name the
Tapajos is known for its upper half. To this unknown river Colonel
Rondon had given the name Ananas, because when he came across it he
found a deserted Indian field with pineapples, which the hungry
explorers ate greedily. Among the things the colonel and I hoped to
accomplish on the trip was to do a little work in clearing up one or
the other of these two doubtful geographical points, and thereby to
push a little forward the knowledge of this region. Originally, as
described in the first chapter, my trip was undertaken primarily in
the interest of the American Museum of Natural History of New York, to