48 ROYAL SOCIETY OF CANADA 



an auxiliary outlet for use in adjusting the head. There is also a small 

 :^-inch tube, with a cock, which is useful for tine adjustment, in case the 

 inflow while running is slightly in excess of the discharge. 



It is necessary, in experiments of this kind, to have means of aecu- 

 ratel}- fixing the head, and the arrangements provided for its observation 

 and regulation will now be noticed. The tank is provided with a gauge- 

 glass extending from the top to the bottom. A brass scale is fitted 

 alongside this glass, and this is graduated in feet above the centre of the 

 orifice, which is arranged to occupy the same position in every case. To 

 assist in fixing the level of water in the tube, a carrier is arranged to 

 slide up and down the brass scale ; this carries a horizontal wire. There 

 is also a mirror behind the glass tube. The height of water is fixed by 

 getting the surface of thi' water, the cross wnre and its reflection in the 

 mirror and any given mark on the scale in the same line. In order to 



• observe the level of the water in this glass gauge, it is necessary for the 

 •observer to climb up a ladder to the level of the water. To obviate the 



necessity of doing this constantly there is an additional indicator, consist- 

 ing of a float on the surface of the water, with a cord attached, which 

 passes over a large light pulley at the top of the tank, and has a weight 

 \^t the other end, on the outside, to keep the cord taut. A pointer is 

 arranged friction tight on this coi'd, so that whatever may be the height 

 of water in the tank, the pointer can alvvaj's be adjusted to ])oint to a line 

 on a brass plate placed in a convenient position on the side of the tank. 



The method of adjusting the head to any required level is by means 

 of a three-wa}' valve, which will either let in water to the tank or allow 

 it to esea])e. This valve is provided with a long vertical spindle with a 

 inumber of handles, one of which can be reached from any ])Osition on 

 the lailder beside the gla.ss gauge. By this means the observer, standing 

 in a position to see the height of the water in the gauge, can let in or let 

 •out enough water to bring the level to ;iny desii*ed point. This is done, 

 previously to each experiment, while the oriflce is closed. The pointer 



• on the cord is then made to ])oint exactly to the line on the brass ])late. 

 During the experiment the head is kept con.stant by means of this pointer, 

 which can be watched while within reach of the inlet and outlet valves. 

 This Hoat and cord indicator shows very small variations in the head. It 

 is liable to a certain amount of error owing to the varying length of the 

 cord and other cau.ses ; so the precaution has to be taken of frequently 

 .checking it by means of the glass gauge. 



An important consideration in connection with such experiments as 

 these is the pn-paration of the orifices. The actual area of the oriflce is 

 somewhat difficult to establish within a small percentage of error, on 

 accoinit of the small size of the orifices themselves. 



The orifice plates themselves are made of gun-metal. The centre of 

 tlie orifice coincides with that of the platt-. The orifices are worked 



