H. S. I. SMAIL. CXCI. 
dials when the brake is again put on, neither of which 
assumptions is true. 
GAUGING MURRUMBIDGEE RIVER. 
Method (b) is the most exact method when the water 
level is fluctuating rapidly, and is being made use of at the 
present time by the author at Gundagai, where the Mur- 
umbidgee is rising at the rate of 8 or 9 feet per day. This 
consists of giving the meter arun of 500 revolutions at 
one point on each vertical, as no matter whether the sur- 
face velocity be the greatest or whether the maximum 
velocity be below the surface, a point at which the velocity 
is equal to the mean remains remarkably constant at a 
certain point below the surface at each vertical. Hum- 
phreys and Abbot give this point as °63 depth at each 
vertical. T.G. Hllis gives °64 depth, and Wheeler and 
Lynch give °67 depth. The Cornell experiments by E. C. 
Murphy agree with the latter, but they as well as those by 
Wheeler and Lynch were made in flumes or channels having 
a greater ratio of depth to width than on most rivers. 
From a consideration of some 60 vertical velocity curves 
on the Sydney Water Supply Canals (upper and lower), 
