President's Address. 67 



tainly is in the direction of causing the river to break over its banks 

 up stream from the dam. This influence, however, is not so en- 

 tirely an unmitigated evil as is an ordinary obstruction in a river 

 channel. A tree or bridge pier or snag checks the velocity of water 

 and in no degree works a recompense for it, so that its influence is 

 entirely bad. A dam across a water stream checks the velocity of 

 water, but in turn necessarily in a measure compensates for it by 

 causing the water to flow with greater velocity as it passes the dam. 

 Observations and measurements universally show that when water 

 begins rising in a stream it rises faster below the dam than it does 

 above, so that when a stream becomes sufficiently high the presence 

 of the dam cannot be told from the appearance of the surface of the 

 water. 



Let us suppose that we have a dam five feet high. At low stages 

 of water the water-level above the dam will be five feet above the 

 water-level below. As a flood approaches this difl^erence in level 

 begins diminishing — first to four and one-half, then four feet, 

 then three and one-half, etc. A little thought will show why this 

 is so. Let us assume that we have a rise of ten feet above the 

 dam, so that a volume of water ten feet in depth will flow over the 

 dam. Now, every one knows that the water at and near the sur- 

 face flows very much faster than water near the bottom of the 

 stream. The farther the surface is removed, therefore, from the 

 top of the dam as the flood rises the less influence the dam will 

 have on this rapidly flowing water, which constitutes the main 

 run-oflp of the stream. During low stages the dam, therefore, pro- 

 duces its maximum effect, which effect is gradually decreasing in 

 percentage proportion as the water rises. 



In confirmation of the above theoretical consideration it should 

 be added that a careful investigation of the Kansas river valley im- 

 mediately after the fiood of 1903 failed to reveal any material dif- 

 ference in damages produced along the river valley above and below 

 the dam at Lawrence. I had eight assistants working under me 

 examining the river valley from its mouth westward to Junction 

 City. These were young men not one of whom was a citizen of 

 Douglas county. One of the charges I gave them was to look care- 

 fully into this question and deter^nine if possible whether or not 

 the dam at Lawrence produced any appreciable effect in the volume 

 of damage done by the flood below the dam and above the dam. 

 The unanimous statement was that, so far as they could determine 

 by careful observations, flood damages were just as great below as 

 above. 



