886 EXPEKIMENT STATION RECORD. [Vol. 35 



Report of the undersecretary of state on the Ministry of Public Works, 

 1914-15 (Rpt. Min. Pub. Works Egypt, 1914-15, pp. VII +126, pis. 21).— This 

 report deals especially with irrigation and irrigation works in Egypt. 



The flow of water over sharp -edged notches and weirs, H. J. F. Gousley 

 and B. S. Crimp (Minutes Proa Inst > -Civ. Engin. [England], 200 {1915), pt. 2, 

 pp. 388-408, figs. 4). — Experiments on tho flow of water over sharp-edged, tri- 

 angular, and rectangular weir notches are reported. 



It was found that " the flow over a triangular notch is proportional to E^-*^ 

 and varies in direct proportion to the ratio of width to height. The general 

 law is Q=2.48, n. W", in which Q is the discharge in cubic feet per second, 

 n the tangent of half the included angle of the notch, and H the head in feet. 



" The flow over any trapezoidal notch is equal to the flow over a rectangular 

 weir of equal length with two end contractions plus the flow through a tri- 

 angular notch of corresponding angle. 



" The flow over a rectangular weir with end contractions varies as B}-" 

 and increases rather more rapidly than the length, i. e., as L^•''^ and is given 

 by the formula 0=3.10 L'-°' H"^", which applies to all weirs up to at least 19 feet 

 in length, and, to judge from the comparisons made for short weirs, for heads 

 up to half the length of the weir, provided the depth of pool below the sill of 

 the weir is not less than twice the head. In the formula, Q is cubic feet per 

 second, L the length in feet, and H the head in feet." 



Abnormal coefficients of the Venturi meter, A. H. Gibson {Minutes Proc. 

 Inst. Civ. Engin. [England], 199 {1915), pt. 1, pp. 391-408, pi. i ) .—Investiga- 

 tions conducted at University College, Dundee, on the causes of abnormal 

 values of the coefficient C used in the base formula for the venturi meter, 



Q = CA^r^ are reported which led to the following conclusions: 



V TO^ — 1 



" In any meter, in good order and of normal proportions, friction does not 

 affect the value of C by more than about 2 per cent so long as the diameter of 

 the pipe line is greater than about 2 in. For large meters at velocities exceed- 

 ing 1 ft. per second the effect is in general less than 1 per cent. The effect 

 increases very slightly with a diminution in velocity. With pipe line velocities 

 less than about 0.5 ft. per second the steadying of the velocity at the throat 

 causes a distribution of velocity under which the kinetic energy at the throat 



v' 

 is appreciably greater than feet. In a meter fitted with the usual U-tube 



gage, or any modification of this type, this may reduce the apparent value of 



C to a minimum of about 0.75 at low velocities. Values of C obtained from a 



meter when measuring a pulsating flow are less than with constant flow. If fc 



is the proportional fluctuation of velocity per cycle on each side of the mean, C 



I 1? 

 is reduced in the ratio l-s-yi-f-. Except, as where used for metering the 



discharge from a reciprocating pump, fc may be large, any such effect is in 

 general small. 



"The effect of whirl in the water approaching the meter is to increase the 

 value of 0. The effect is, however, small; is approximately constant at all 

 velocities ; and it is extremely improbable that in the normal pipe line it exceeds 

 about 1 per cent. 



" Where the pressure orifice at throat or entrance of a meter consists of a 

 circumferential gap, the width of this gap has some effect on the value of C. 

 In a meter having a throat diameter of 1 in. the effect is, however, small, 

 except at very low velocities, so long as the width does not exceed 0.4 in. The 

 author found that at medium and high velocities maximum values of C were 



