24 BULLETIN 832, TJ. S. DEPARTMENT OF AGRICULTURE. 



The values obtained for n are shown in Table 2, and, as would be 

 expected from the condition of the channel, are even higher than those 

 found for a bankful stage of Bogue Hasty. Some trouble was ex- 

 perienced during the measurements, due to drift collected at the 

 bridge, and this may be in part responsible for the rather high values 

 obtained for n. (See fig. 3 E for average cross section.) 



DISCUSSION OF BOLIVAR COUNTY EXPERIMENTS. 



From the foregoing description of .the results obtained and the 

 conditions existing in each of the canals it is apparent that the most 

 dependable data were obtained for the Bogue Phalia and Bogue 

 Hasty channels. These results show that a value of n of about 

 0.030, or slightly larger, applies to channels similar to the whole 

 channel of Bogue Phalia. 



The values obtained for the whole channel of Bogue Hasty, Pecan 

 Bayou, and East and West Bogue Hasty demonstrate the effect of 

 the conditions described upon the value of n. These values can be 

 used as a guide in the design of drainage channels where like condi- 

 tions can reasonably be expected to exist, due to the lack of mainte- 

 nance of the channels or to the caving in of the banks. These data 

 ought also to serve a useful purpose in estimating the discharge of 

 old channels where like conditions prevail. 



EXPERIMENTS IN WESTERN TENNESSEE. 



These experiments were conducted during the years 1916 and 1917. 

 Values of n were obtained for six different courses of the channel of 

 the South Forked Deer River and for courses of the channels of the 

 North Forked Deer River, Huggins Creek, Sugar Creek, and Cypress 

 Creek. The author was assisted in these experiments by H. S. 

 Andrews and A. L. Lane, junior drainage engineers of the bureau. 



SOUTH FORKED DEER RIVER NEAR ROBERTS. 



Gaugings of this channel were made from a single-span highway 

 bridge about 1 mile south of Roberts. The slope course was 1,412 

 feet long, and was located just above the gauging station. The channel 

 increased gradually in size from the upper to the lower end of the 

 course, there being no abrupt changes in size or shape along the 

 course (figs. 5 A, and QA). There was no vegetation nor any marked 

 irregularities in the channel (Plate VII, fig. 2). Considerable erosion 

 had taken place in this channel to which was largely due, no doubt, 

 the remarkably few irregularities and the freedom from vegetation. 

 As shown in Table 3, lower values of n were obtained for this channel 

 than for any of the other channels in Tennessee. 



