108 HarpPER: Trip ON WARRIOR AND TOMBIGBEE RIVERS 
surveys in the southeastern states), a young minister (who was also 
a conchologist), a navigator, and a negro cook. 
The trip was made in an old house-boat refitted for the occa- 
sion, towed by a 20-horse-power gasolene launch. Our best speed 
(including the acceleration of the current of something like a mile 
an hour) was about six miles an hour; and the distance covered 
between sunrise and nightfall each day averaged about twenty-six 
miles, but varied greatly with the number of bluffs to be examined 
and the difficulties encountered, from one to forty-nine miles. 
Our voyage began on the morning of October 7, 1908, at the 
fall-line on the Warrior River, in the city of Tuscaloosa, and 
ended on the evening of the 16th at Jackson, near the head of 
tide-water on the Tombigbee, 261 miles from Tuscaloosa and 100 
from Mobile, by the river. Although there would doubtless have 
been much of botanical interest below Jackson,* the geologists did 
not care to go any farther than that, for there are said to be no 
important bluffs on the tidal part of this river. The river at this 
time was almost at its lowest stage, because its headwaters are 
mostly in regions which have their greatest rainfall in winter and 
spring and least in summer and fall, as seems to be the case in 
most of the country between the mountains and the coastal plain 
in the southeastern states. | This state of affairs greatly facilitated 
our examination of the banks, but at the same time it caused most 
vexatious delays at the shallowest places, which were usually just 
below the mouths of swift creeks, which bring down coarse sand 
and gravel faster than the slower current of the river can remove 
it, and thus form shoals. 
But on the whole probably no better season of the year could 
description of some of the swamps between Jackson and Mobile can be found 
in Bartram’s Travels, Ae 35 seg 6; and parts of the same have 
some of Dr. Mohr’ itings. 
been described in 
rte region now under consideration there is hardly 
any previous Setaniedt ecu: to be cited. Excellent geological descriptions, with 
cuts of some of the bluffs, by Dr. Smith and others, have been published in Bulletin 43 
of the U. S. Geological Survey, 1887, and in the ** Report on the geology of the 
coastal plain of Alabama,’’ issued by the state survey in 1894 
he surface of the Warrior River at Tuscaloosa is about go feet above sea-level 
at low water, usually in October, and 150 feet at high water, usuallyin March. (Far- 
ther down stream the elevations are of course Ser tg lower.) The smallest 
flow ever recorded there was 90 cubic feet a seco ctober, 15, 1897, and the 
mae 136,687 cubic feet, on April 18 Ig00; an enormous variation. The average is 
ut 8,000, which is nearly two cable feet a second for every square mile drained. 
