H. S. I. SMAIL. CLXXXVII. 
It is partly to this fact, in the author’s opinion, that the 
individual gaugings in most cases le so irregularly about 
the mean rating curve for each gauge, as the individual 
current meter gaugings are much more regular. It is im- 
possible to entirely get rid of the action of the wind. On 
irregular beds such as most natural streams have, the rod 
must be short enough to pass over the shallowest part of 
its run, and may therefore and indeed in the majority of 
cases is too short for the greater part of the run, therefore 
the velocity as given by the rods is too great. From a 
comparison with current meter gaugings the author has 
come to the conclusion that the gaugings taken on the 
New South Wales rivers with rod floats are about ten per 
cent. too high. 
Current Meters.—By far the most accurate and speedy 
method of gauging a natural or artificial stream is by means 
of the current meter, provided the meter is of good pattern 
and has an electrical connection registering the number of 
revolutions above the water surface. Murphy says,’ ‘*The 
meter without some electrical device for indicating to the 
observer the revolutions in a given time, and which must 
be lifted out of the water to read the revolutions, is a thing 
of the past. Too much time is lost in making the readings. 
The acoustic meter is not a success, the clicks are not 
sufficiently loud to be heard where the water is deep and 
noisy.”’ 
The meter adopted by the Public Works Department is 
one made by Amsler-Laffon registering each fifty revolutions 
by ringing an electric bell. The author has had one of 
these meters in use for about 18 months, and although used 
in very dirty and muddy water for the most part, it has 
altered very little in rating since it was first rated. The 
* Water Supply and Irrigation Papers of the U.S. Geological Survey, 
No. 64, p. 22. 
