FLOW OF Y/ATER IN DREDGED DRAINAGE DITCHES. 7 



The increase in the value of n, due to the growth of grass in channels, 

 is shown by the results of the experiments made in North Carolina 

 (Table 5). These experiments were made during the winter and the 

 following summer, before and after the growth of grass in the channels. 



. DESCRIPTION OF CHANNELS. 



At the right-hand side of each of the Tables 1 to 6 is given a complete 

 description of the drainage channel. These detailed descriptions, 

 supplemented by the views of the channels shown in the plates, are 

 intended to assist the engineer in the proper choice of n for his partic- 

 ular use. Under the description of channels are included the factors 

 or conditions that influence the now of water in a channel. These 

 factors are described under the following headings: Course, cross 

 section, side slopes, bottom, soil, and condition. Also the approxi- 

 mate date when the channel was constructed is given, from which can 

 be determined the age of the channel at the time the experiments 

 were made. 



COURSE. 



The length and alignment of the course of channel are discussed 

 under this heading. Where the length of the course is unduly short, 

 the probability of error in the slope measurements is somewhat 

 greater than where it is comparatively long. It was impossible in 

 many instances, other governing factors being duly considered, to 

 follow the general rule adopted, namely, that the length of the slope 

 course should be at least 10 times the top width of the channel. 

 With the exception of two courses of channels, the alignment of 

 the slope courses was practically straight. These exceptions consisted 

 of a bend in the Sugar Creek channel and of several irregular crooks 

 in the old river channel at Campbell's levee, near Jackson, Term. 

 (Table 3). 



CROSS SECTION. 



Under this heading variations in the shape of the cross section 

 along the courses are noted, and the reader is referred to figures 2, 4, 

 6, 8, 10, and 12, which show the per cent variation from the average 

 cross-sectional area for all cross sections along the slope courses. In 

 most cases this per cent variation is shown for low, medium, and 

 high stages. From these figures an idea can be obtained as to the 

 progressive changes in size of the channel along the courses and 

 whether these changes are gradual or abrupt. For instance, in figure 

 2 A, it is seen that the per cent variations are much larger and the 

 changes in size of the channel along the course are much more 

 abrupt for the low than for the high stage, which accounts largely 

 for the higher values of n obtained for the low stages as recorded in 

 Table 1. 



