FLOW OF WATER IN CHANNELS AND PIPES. 49 



inserted at such a level that it will run about half 

 full when constituting the ONLY outlet for the im- 

 prisoned water. The enquirer can then form a toler- 

 ably reliable estimate of the flow by counting the 

 number of bucketfulls of water collected, say in a 

 minute, and knowing the capacity of the bucket. It 

 is only applicable to very small flows ; somewhat 

 larger flows may be measured by allowing the pipe 

 to run full force, and measuring the depth from still 

 water level to the centre of the pipe. This consti- 

 tutes the "head" and with the diameter of the pipe 

 gives a means of calculating the probable discharge. 

 We shall discuss flow in pipes later. It will nearly 

 always be the case, however, for conditions 2 and 3 

 that a proper gauge will have to be erected across 

 the stream, and the use and construction of such a 

 gauge as the writer uses will now be described. It 

 may be well to note, however, that where the stream 

 is sluggish and is fairly deep and uniform in cross 

 section, a tolerable estimate 

 of flow can be made by 

 ascertaining the velocities. 

 For instance, take such a 

 case as fig. 39, which might 

 be a mill race or other artifi- 

 cial channel. A few experi- FlG - 39- 

 ments with a float will give the velocity at the centre 

 between two known points. An orange or gutta- 

 percha ball is recommended, or a rod float consisting 

 of a light wooden rod loaded with lead at the bottom 

 and with a piece of cork at top will be found very useful. 

 They should be of such a length that the weighted 

 portion is a few inches from the bottom. With this 



4 



