-months' amount to a two-months' quantity, though they had not
yet been on the way quite sixty days; that is, they had used up eight
months' supplies in two months, including a mountain sheep and a deer
the hunters had brought down, and they were barely more than half-way
to the end of the journey. At this alarming rate they would be starving
long before they saw the walls of the Grand Canyon break away.

Nevertheless no thought of pursuing any course but the one planned
occurred to them, and on July 21st they cast off from the sand-banks and
were carried rapidly down on the swift torrent of the Great Colorado.
They had not gone far before plenty hard work was furnished, in the
shape of two portages were necessary to pass particularly dangerous
places, and numerous bad rapids to run. In the afternoon the Emma Dean,
in attempting to navigate one of the more favourable-looking foaming
descents, was swamped, pitching Powell and the others headlong into the
roaring flood. They were fortunately able to cling to the boat till they
floated into more tranquil waters, where they managed to climb on board,
signalling the other boats to land before the plunge. This they could
do, and the boats were brought down by a portage, which took all the
rest of the day. The approach of darkness compelled a halt for the
night on some rocks where they had barely room enough to lie down. Three
much-needed oars had been lost with the capsize of the Dean. These were
sadly missed in the rough water that surrounded them the following day,
so at the first large pile of driftwood they made a landing and secured
a cottonwood log for oar-timber. While the oars were making, Powell and
his brother climbed up to where some pinyon trees were seen growing,
and collected a quantity of gum with which to calk the leaky boats.
They needed all the preparation possible, for the rapids now came ever
thicker, ever faster, and more violent. The walls also grew in altitude
from the thirteen hundred feet of the Junction to fifteen hundred feet,
then to eighteen hundred feet, nearly vertical in places.

An examination of the barometric record was now made to see how much
they had by this time descended toward sea-level, and, by comparison,
about what might be expected in the river below. The conclusion was
that though great descents were still ahead, if the fall should be
distributed in rapids and short drops, as it had been above, and not
concentrated great plunges,