30 PROCEEDINGS COTTESWOLD CLUB 1915 
I regard as the first thing to be done in regulating the course 
of rivers, the fixing of the low-water channels. This being 
done, the river may still overflow and inundate the district, 
but, as the flood subsides, it will again restrict itself to the 
proper channel. 
I may not now discuss at length the development of rivers. 
Enough to say, first, that I cannot concur in accepting the 
generally-accepted theory of “ primary consequent, of sub- 
sequent and of obsequent streams,” and, secondly, that to 
me a river system, a principal stream having tributaries flowing 
in on either side, is the outcome of a process of evolution. It 
has been evolved out of a network of channels. In this, water, 
at any point, would have alternative routes of possible flow. 
These routes may, for a time, be equally free ; but one or 
other of the streams will, sooner or later, cut down its channel 
more deeply then any other, and this one will be adopted, 
although its course may not be the most direct, and even 
although it be so circuitous as to require the stream to go back 
upon itself and reverse its course. The disused routes, no 
longer cut down by water flowing through them, may be 
effaced or may remain as valleys, drained in opposite directions, 
with a stream at each end. These valleys serve as passes 
between river systems ; they have often been utilized for roads, 
canals and railways. 
Rivers generally have a winding course. “ In serpent 
error rivers flow ; ” and, sometimes they seem to be capricious 
in their ways. I, however, hold firmly to the faith that rivers 
are always under the influence of the constant laws which 
govern the flow of water. Foremost among these is one that, 
when flowing to a relatively lower level, it always has 
a tendency to flow in a straight line, in accordance with the 
force of gravity. Rivers have no "natural tendency to wander”: 
the direction of the flow, at every part of a winding course, 
is a resultant between the effect of diverting influences and the 
tendency to flow in a straight line. A river may be diverted by 
a change in the slope of the land, when the force of gravity 
will operate in a new line : it may be pushed aside by an 
obstruction, or it may be drawn aside by the attracting in- 
fluence of another stream. This last cause is not generally 
