FLOW OF WATER IN DREDGED DRAINAGE DITCHES. 51 



10 E) There were some roots and dead vegetation in the channel 

 during the winter months. Silting had occurred along the course, and 

 there was a mud island 3 feet wide by 10 feet long lying lengthwise of 

 the channel a short distance above the lower end of the slope course. 

 During the summer months the island was covered with water grass. 

 There was also a growth of grass along the edge of low-water stages 

 and two patches of grass in the channel along the course. 



The highest stages for which values of n were determined were 

 about half -bankf ul. Referring to Table 5, it can be seen that the 

 values of n obtained during the summer months are higher than those 

 obtained during the winter months, owing to the growth of grass in 

 the channel; also that the values for the summer months increase as 

 the stage decreases, due no doubt to the fact that for the lower stages 

 the grass rilled a much greater proportion of the cross-sectional area 

 of the channel. Plate XVIII, figure 1, is a view of the channel, and 

 figure 9 E shows the average cross section. 



DISCUSSION OP NORTH CAROLINA EXPERIMENTS. 



Attention is especially called to the fact that in these experiments 

 values of n were not obtained for bankf ul stages, and this should 

 be kept in mind when comparing the views and descriptions of the 

 channels with the values of n obtained. A value of n for a near- 

 bankful stage was obtained on the Back Swamp channel, but in the 

 other channels the highest stages for which values of n were deter- 

 mined were about half, and in some instances considerably less than 

 half, of a bankful stage. 



The chief value of these experiments lies in the fact that values of 

 n were determined before and after the growth of grass in three of 

 the channels, from which a general idea can be obtained as to the re- 

 tarding effect of grass upon the flow in water courses. 



With the exception of Back Swamp the discharges of the channels 

 were quite small, and for this reason it is not believed that the results 

 are as accurate as were obtained for the other experiments herein 

 discussed, since the effect of errors in making small discharge measure- 

 ments is usually greater than in the measurement of large discharges. 



EXPERIMENTS IN FLORIDA. 



These experiments were made on the main channels of the Fells- 

 mere, Vero, and Fort Pierce Drainage Districts, and on two courses 

 of Lateral No. 2 of the Fort Pierce Drainage District. They were 

 conducted by F. E. Staebner, drainage engineer of this bureau, 

 during 1916 and 1917. 



