164 KANSAS UNIVERSITY QUARTERh\'. 



From a quarter to half a mile above the bridge there is a bend in 

 the river, which deflects the water to the north bank at the bridge, 

 making the maximum velocity on the north side of the north pier, 

 and causing considerable edd3ang on the north side of piers Nos. 3 

 and 4, when the water is high. 



The bed of the river at this place is rock from the south abut- 

 ment to a little north of pier No. 3; the rest of it is sand. The 

 south half of the dam is cement masonry on rock, through which 

 there is little seepage; the north half is made of timber cribs filled 

 with rock and through these and the sandy bed there is consider- 

 able seepage. The discharge measurements were made during the 

 months of July, Aug., Sept., and Oct., and are numbered i, 2, 3 

 and 4, and c in the order in which they were made. These measure- 

 ments expressed in cubic feet per second arc plotted in Fig. 3, as 

 abscissas using the corresponding gauge readings as ordinates. 



Three current meters were used in measuring the velocity. 

 Measurement No. i, was made with Haskell meter No. 6, No. 2, 

 with Price meter No. 19, and the other eight with Haskell meter 

 No. 17. The depth was measured every five feet, and the velocity 

 every five to twenty-five feet depending on the rate of its change. 



Section C D being large it is impossible to measure accurately 

 the velocity when the gauge reading is much less than one foot, so 

 that measurements No. 9 and 10 were made at Lecompton, 11 

 miles up the river on the Santa Fe R. R. The only tributar}- of 

 the Kansas river of any size between Lawrence and Lecompton is 

 the Delaware. The average discharge of this stream on Oct. 25, 

 when measurement No. 10 was made, was less than 12 cubic feet 

 per second. Fig. 4 shows the Lecompton section on Oct. 5, when 

 measurement No. g, was made. 



I have computed the discharge from Mr. Bowersock's record of 

 mean weekly gauge readings by two methods. Tor the period from 

 Jan. 1, 1891, to July 31, 1S65, I have found tlic probable mean 

 daily gauge readings by plotting the weekly means and drawing a 

 curve similar to that for the months Aug. to Nov. '95, and so that 

 the mean of the seven daily ordinates of this curve is the same as 

 Mr. Bowersock's weekly mean. Entering the rating table w-ith 

 these probable daily mean gauge readings the discharge for each 

 day was found. For the period Jan. i, 1886 to Dec. 31, 1890, the 

 discharge for each week was found by entering the I'ating table with 

 the weekly mean gauge reading and multiplying the corresponding 

 discharge by seven. The rating table gives the discharge in acre feet 

 per 24 hours. It is computed from the rating curve, Fig. 3, by re- 

 ducing the discharge in cubic feet per second, to acre feet per 24 



