424 REPORTS ON THE STATE OF SCIENCE, ETC. 
47, THE CONTINUOUS RECORD OF WATER LEVELS CONVERTED INTO 
FLow REcorDs. 
There is, firstly, the table of flow for each } in. or so of water level, which 
has resulted from the river gaugings. 
From the graph of the water level recorder, the average water level of 
each 3 hours is tabulated on the main records table and each 3-hourly 
value is converted into flow. ‘The daily average flow is then tabulated. 
On special floods, shorter time intervals may be used on a similar table. 
Such tables will provide all necessary data for any investigation required. 
The essential point to make clear is that the records should be, firstly, 
records of water level and the conversion to flow should not be made on 
the water level recorder graph. The reasons for this decision are many. 
The shortest way of putting the point is that flow measurements are sub- 
sidiary to water-level measurements and that the connection between the 
two may change (1) by changes in the river, and (2) by improved measure- 
ments. But beyond that, water levels are a definite record of storage and 
stage throughout the whole river system. 
In further reference to river gauging, extracts from Prof. Dixon’s and 
Mr. Griffiths’ statements are attached (see Appendices a and b). 
APPENDIX F (a). 
NOTES ON THE GAUGING OF THE RIVER SEVERN, 
AT BEWDLEY. 
By S. M. Dixon. 
1. The gauging site, half a mile above Bewdley Bridge, is in a section of 
the river which is sensibly straight and of reasonably uniform width for a 
little over three-quarters of a mile. About a quarter of a mile above the site is 
a short section of rapids, which dries partially at extreme low flows. Another 
similar shallow section occurs three-quarters of a mile below the gauging site. 
The river, near the site, flows between high steep banks, its width varying 
from about 140 ft., at minimum flow, to about 180 ft., at which point it over- 
flows its banks into level fields on either side. ‘The bottom is of rock, and 
the depth at minimum flow varies from 7 ft. in midstream to about 3 ft. 
close to the banks. 
There is a gauge post at the site, consisting of four cast iron sections, 
bolted to steel channels, set in concrete on the bank, each projecting 3 ft. 
from the ground, and recording over a total rise of 11 ft. in water level. 
Each foot is divided into tenths. 
Daily records of the water level are read on a permanent cast iron scale 
fixed just below Bewdley Bridge, about half amile down-stream of the gauging 
site, and on a sloping concrete scale cut in the Aqueduct carrying the 
Birmingham Corporation’s water supply across the river two miles above 
the site. An automatic recording gauge is also installed at the latter place 
and is attended to weekly by the Corporation Water Department’s walksman, 
who lives close by. The sloping concrete scale serves to check the setting 
of the automatic recorder. 
The zero of the gauge at Bewdley Bridge was chosen to correspond with 
the lowest water level which was likely to occur. The zeros of the other 
fixed gauges were chosen arbitrarily, the zero of the automatic gauge at the 
