540 DISCUSSION: FORESTS, RESERVOIRS, AND STREAM FLOW 
Mr. Chitten- the mountains leaves bare ridges of sterile rock, and the rich organic 
den. 
mould which covered them, now swept down into the dank low grounds, 
promotes a luxuriance of aquatic vegetation that breeds fever, and 
more insidious forms of mortal disease, by its decay, and thus the 
earth is rendered no longer fit for the habitation of man.” 
Oblivious to the humor of the situation, Marsh adds: “To the gen- 
eral truth of this sad picture there are many exceptions.” 
II. 
The writer regards the following extract from an address by 
Colonel Thomas P. Roberis before the American Forestry Congress, in 
September, 1885, as well worth careful study. It is the first rational 
explanation that has come to his notice of the fact that the low-water 
flow of our rivers is not less than it used to be, in spite of the gen- 
erally accepted fact that the little springs and rivulets dry up more 
than formerly. Colonel Roberts’ life-long experience upon the Ohio 
River, and his inherited fund of information upon the subjects, through 
his father, the eminent hydraulic engineer, W. Milnor Roberts, whose 
career was also largely identified with that stream, give his opinion 
upon the matter, considered in the abstract, more than ordinary 
weight: 
“A river of large drainage area is made up, to be sure, from a mul- 
titude of tributaries. Over an area of 20000 square miles, which is 
the basin of the Ohio at Pittsburg, there are, even in the seasons of 
the most protracted droughts, almost daily thunder-storms, of limited 
area, perhaps, but each deluging in turn one or more of the minor 
tributary valleys and causing local freshets. No rainfall may be 
observed at any designated station for two or more weeks at a time; 
nevertheless the river represents the aggregate of their effect, its low- 
water discharge being in reality made up of numerous miniature 
freshets. In other words, I argue that the low-water discharge of the 
Ohio River is, to a very great extent, maintained by a circulating 
series of minor freshets. 
“A dry-weather spring yielding two gallons of water per min- 
ute, making seventy-five barrels daily, would be accounted enough 
even for the maintenance of a large stock farm. The lowest stage 
of water ever known upon the upper Ohio was in 1838, when for sev- 
eral days the least discharge of the river was 552000 gallons per 
minute, to maintain which it would require the united tribute of 
276 000 such springs. For the average season of drought it would re- 
quire more than 500000 such springs, to maintain the low-water dis- 
charge of the Ohio River. On the other hand, it would require that 
only ninety-seven square miles of the entire territory of 20000 
square miles be daily visited with a storm yielding for discharge a 
little over half an inch of rainfall. 
“These figures are presented merely to set forth very briefly how 
great rivers may maintain their low-water discharge. I am persuaded 
