14 BULLETIN 194, U. S. DEPARTMENT OF AGRICULTURE. 



actual time to the tenth of a second being taken for 20, 30, or 40 

 revolutions, or whatever the number might be. This method of 

 operating a meter is better than the old one of running the meter for 

 a given number of seconds, thereby introducing an error due to 

 fractional revolutions. 



Adjustable Tiook gauge. — ^At some point within sight of the man 

 operating the meter a small hook gauge was set. Its range was about 

 0.6 foot with a vernier reading to thousandths of a foot. This gauge 

 was used to determine the rise or fall of the water surface in the canal 

 during the measurement. The final accepted elevation of the water 

 surface was the mean elevation throughout the meter measurement. 

 All soundings for cross section and the' final setting of the bench 

 marks from which the hydraulic grade was determined were based on 

 this mean value of the elevation of the water surface. 



Hydraulic grade. — Hydraulic grade was determined during the 

 latter part of or after the measurements for discharge. In earthen 

 channels stakes were firmly driven at each end of the reach chosen 

 until the top of the stake was about 0.07 foot below the surface. 

 After the wire nails were carefully set, as described on page 10, a 

 line of levels was riui between the bench marks at the ends of the 

 reach. The level was kept in careful adjustment, and in addition 

 all instrumental errors were obviated by equalizing the sights. 

 This line of levels was as a rule determined by one sight in each direc- 

 tion. A clear-cut target was set very carefully, and the vernier read 

 by both levelman and rodman, independently. Numerous tests 

 showed that the differences obtained in the total fall between the ends 

 of a reach were well within the allowable error for careful leveling. 



Data from which to determine the average value of the hydraulic 

 mean depth, comprising sufficient cross sections taken throughout 

 the length of the reach tested, were taken as a rule by means of a tape 

 and level readings on a rod held at the bottom of the channel at suf- 

 ficient verticals to give true values of the area and the wetted per- 

 imeter. It was obviously impossible to drive stakes and set nails 

 vertically in the tops thereof in flumes and concrete-lined sections. 

 In these sections wire nails were driven horizontally into the bank 

 at the water surface, the head of the nail forming a knife edge which 

 quite clearly defined the surface of the water. They may be protected 

 by stilling boxes in low velocities, and in high velocities the fall is so 

 great that refinement in levels is neither practicable nor necessary. 

 In concrete linings the ends of the reach were so chosen that they 

 came at expansion joints, where it was not a difficult matter to drive 

 a nail or, in cases where the lining formed an unbroken surface, a 

 small hole was first drilled with a machinist's centering punch and 

 the wire nail set in this hole. 



