DISCUSSION: FORESTS, RESERVOIRS, AND STREAM FLOW 439 
From the Fifties to the early Nineties the lumbering operations Mr. Labelle. 
on the Susquehanna were very brisk, and enormous quantities of 
square timber were taken down to tide-water. Many mills, especially 
in the Williamsport district on the West Branch of the Susquehanna, 
have also been busy transforming the forests into lumber. The flood 
of June 2d, 1889, which coincided with the flood at Johnstown, Pa., 
brought to tide-water millions of logs, which occurrence made clear 
the vast extent of lumber cutting in progress in the upper part of the 
water-shed. Since 1895 square timber rafting has ceased on the river, 
but the manufacturing of lumber and the ‘consequent deforestation 
are still progressing at a rapid rate. 
As to the Schuylkill River, E. F. Smith, M. Am. Soc OC. E.,, 
General Manager of the Schuylkill Navigation Company since 1865, 
states that little or no deforestation has taken place in the water-shed 
during the last 40 years, and that the Schuylkill valley is largely a 
farming and not a densely wooded district. 
As the low water at Harrisburg, on the Susquehanna, has not been 
influenced by engineering works, the average low water for the last 
two decades (1891-1908) has been introduced in Table 15. 
Bearing in mind that deforestation has progressed in all the water- 
sheds except the Schuylkill from the early decades to the present 
time, it will be seen from Table 15 that the floods have increased 
in some rivers and decreased in others. In the case of the Ohio, 
at Pittsburg, the mean of the floods (retrogressing from the third to 
the first period and progressing from the third to the sixth) has 
increased in about the same ratio. Comparing the Schuylkill for the 
fourth and fifth decades with the Ohio for the fifth and sixth decades, 
there is a greater increase in the first than in the second case; now, 
if the increase in the former cannot be attributed to deforestation, 
because no deforestation has taken place, why should it be in the 
second case? There is certainly no uniformity in the figures: they tend 
to show that deforestation is not the paramount factor in the causation 
of floods. 
As already stated, the natural features of the low water on the 
Susquehanna have not been disturbed. The zero point of the Harris- 
burg gauge was placed at the low-water mark of 1804. This gauge 
was established in 1890. According to the theory here discussed, and 
after extensive deforestation of the water-shed, low waters should be 
much lower than they were 100 years ago. The fact is that the water 
at Harrisburg went below zero of the gauge only once since 1891 
(—0.04 in 1900); in 1895 it went down to +0.04; all the other 
readings are above 0.16, and the means for the last two decades are as 
shown in Table 15. The low water since 1891 has apparently been 
on the increase. Here also, at first sight, the theory fails to explain 
the facts. The writer says “at first sight” because it is quite evident 
