306 
WEATHER BEREAU RIVER AND FLOOD SYSTEM 
large areas of bottom lands in the ^Mississippi delta. The value 
of the special reports wliich were telegraj^hed at that time b_y the 
Weather Bureau (or Signal Service Office, as it then was) could 
scarce! 3 ' he determined. The\" were the onl\" reports sent directly 
to the i>eople of the flooded districts, and showed daily the com- 
ing rise or fall of the water. A study of these floods showed the 
necessitv of establishing for each of the rivers a certain depth of 
water above which the stages were dangerous to river interests. 
These points were designated as “ danger levels ” and “ danger 
lines,” and were established for the ^Mississippi, Missouri, and 
Ohio rivers during that year. In prosecuting this work, data 
from the l>est available authorities were collected and comjhled 
for the construction of a chart of the basins and watersheds of 
the principal rivers. A river slate was designed, on which were 
outlined the average grades of the beds of the various rivers at 
different jiarts of tlieir courses. The object in preparing this 
chart was to facilitate the tracing of flood-waves and their move- 
ment from one place to another. When an unusualU' heavy 
rain was noted in any watershed, it was known into what rivers 
it must flow and ap|)roximately the rise that would result. A 
knowledge of the rapidiU' with which the flood would travel and 
of the rivers it would pass made it possible not only to follow its 
course, but also to give timely warning of its approach. 
Some idea of the vast destruction of properW bv floods may 
be gathered from the statement that the floods of the spring of 
1881 and of 1882 caused a loss of not less than 615,000,000 to the 
properU’ interests of the Ohio and Mi.ssissipi)i valleys. It may 
also be noted that the flood of the spring of 1882 caused a lo.ss 
of 138 lives in the region from Cairo southward to New Orleans. 
In forecasting stages of water during such flood j)eriods as the 
two mentioned, it must be borne in mind that precipitation may 
be onU' an inconsiderable factor. In those cases vast quantities 
of snow, which had accumulated during the winter, overlay the 
northern states, and with the early rains of spring came abnor- 
mal heat, causing a ver}’ rapid melting of the snow lying over 
manv of the watersheds. In these floods it is ])robable that the 
sudden coming of abnormally high temjieratures was a more 
2 )otent influence than the immediate precipitation. 
The floods of 1884 began in the Ohio valley in Fehruaiy, 
when the river reached the highest stage on record. The Missis- 
sippi river from Cairo to the Gulf also reached a very high stage. 
Am 2 >le and timely warnings were telegraphed to all available 
