442 DISCUSSION : FORESTS, RESERVOIRS, AND STREAM FLOW 
Mr. Labelle. only 838% for the forested areas. From the results Mr. Toumey con- 
cludes that the forests are beneficial, inasmuch as they tend to equalize 
the flow throughout the year. He does not say a word about the 
relative slope and porosity of the water-sheds; the precipitation is 
given, but its character is not mentioned. The influence of the 
different factors producing run-off should have been found before the 
percentage of influence due to forests could have been determined. 
This procedure in the above experiments would have made the results 
more convincing. 
The investigation of the causes of the decreased flow of Rock River, 
Wisconsin, belongs to the second class.* ‘The water-shed has an area 
of about 3 600 sq. miles, is slow-spilling, and has a comparatively high 
retentivity. 
A good proportion of its swampy lands and small lakes has been 
drained; and the custom of laying tile and other drains and of cutting 
ditches prevails throughout the region. This has resulted in a more 
rapid delivery of the rain water into the streams. The rainfall has 
been decreasing since 1885. The forests have decreased from 75% to 
30% of the whole area. Mr. Schwarz, the author of the pamphlet, 
states that the most obvious explanation for the smaller quantity of 
water annually carried by Rock River is the diminution of rainfall; 
he ascribes also a certain amount of influence to the draining opera- 
tions above mentioned. The principal body of water which has been 
drained is the Horicon Marsh, at the upper end of the main branch of 
the river. This marsh was dammed in 1846 for water-power; it has 
an area of 73 sq. miles, and its catchment amounts to approximately 
400 sq. miles, or one-ninth of the water-shed. Later, the dam was 
removed, and the tract was allowed to revert to its original state and 
was put under cultivation. In spite of the conditions just described, 
the pamphlet attributes largely to deforestation the reduction in the 
stream flow. 
It seems to the writer that the conditions obtaining in the water- 
shed were well enough known to permit of making some kind of an 
estimate of the percentage of decrease of run-off due to: 
1—Smaller rainfall; 
2.—Drainage of swamps, low lands, and the Horicon Marsh; 
3.—Increased demand of cultivated lands; 
4.— Deforestation. ; 
The sum of the first three factors subtracted from the total decrease 
in run-off as computed from gauging operations would give the fourth 
factor, or the percentage of decrease due to deforestation. Until this 
computation is made, all estimates of this fourth factor are mere 
speculation. 
*“ The Diminished Flow of Rock River in Wisconsin liinois,” .F, ; 
Bulletin No, 44, Bureau of Forestry. Sale ae 
