‘DISCUSSION : FORESTS, RESERVOIRS, AND STREAM FLOW 413 
with the deforestation. In other words, the one great, predominating 
change in the character of the drainage areas during the period of 
record is deforestation. The history is clear, and the deductions are 
not subject to serious dispute. 
One of the interesting features brought out by these diagrams is 
the remarkable agreement in the progressive trend of the record on 
the Ohio at Wheeling, and those on the Allegheny at Freeport, the 
Monongahela at Lock No. 4, and the Youghiogheny at Confluence. The 
earlier lumbering operations of the Allegheny basin are shown clearly 
by the earlier initial increase in flood tendency. Examples like those 
presented by Figs. 8 to 15, inclusive, might be multiplied several 
times, and they would have been presented were it not for the excessive 
time and labor required in preparing the diagrams. 
There is one point of objection to these diagrams which may 
reasonably be taken: It is that they draw comparison between annual 
precipitation and the number of days of flood. A better comparison 
would apply between the number and depth of flood-producing rains 
and the number of flood days. Such a comparison has been made 
on the Tennessee basin above Chattanooga. No other basins are 
here dealt with in this way, because of the enormous amount of labor 
involved. 
In addition to proving beyond controversy the increase in floods 
due to deforestation, the record is of interest because it shows how 
futile and misleading is the attempt to derive conclusions from flood 
records without taking into account records of precipitation. Inspec- 
tion of the Chattanooga flood record shows a decided decrease in the 
number and duration of floods during the period of record. If this 
were taken alone, it would controvert the forest-control idea. A com- 
parison of the precipitation records on the drainage area, however, 
assures us that the rainfall has decreased, even to a greater rela- 
tive extent than the floods. The final result is that the flood tendencies 
have increased. 
The results for the Tennessee basin cover 24 years, from 1884 to 
1907, inclusive. Although the records of gauge height extend farther 
back than the earlier date, the number of rainfall stations main- 
tained on the basin previous thereto is not sufficient to render safe 
any conclusions with reference to precipitation. It is appreciated 
that a given depth of rainfall precipitated during the winter months 
will generally produce a greater run-off than the same depth would 
produce in August or September. Therefore, an effort has been made 
to eliminate this source of error. Summing up the flood-producing 
rains for the 24-year period, it is found that the total is 335, of which 
3183 occurred from December to May, inclusive, and the remaining 22 
during the other portion of the year. It is apparent that the number 
of such rains from June to November is not sufficient to afford a basis 
Mr. Leighton. 
