84 
Proceedings of the Royal Society 
These three rivers had, therefore, since the completion of the 
training walls, been undoubtedly subject to large deposits of sand ; 
but the original improvement in depth, and increased velocity in 
the propagation of the tidal wave, were still maintained, so that, 
as navigation works, they must be held to have been quite success- 
ful. Indeed, if similar improvements had to be done over again, I 
should not propose to make any change whatever in the mode of 
carrying out the works. It will be seen, however, from these cases, 
that training walls exert a double influence — one directly for good, 
by enlarging and deepening the main channel ; another directly for 
evil, since the deposit of silt behind the walls must reduce the 
scouring power by diminishing the area filled at each tide, and 
thus keeping back a corresponding quantity of tide-water from 
entering the estuary. Thus < the ultimate result for good or for 
evil must be in each case determined by a comparison of the 
relative amounts of these plus and minus quantities. 
Here, then, we have the first and elementary ground of that nega- 
tive answer, which I and others of my professional brethren returned 
to the question proposed for solution in the case of the Mersey. 
If, for instance, the waters of a navigation contained no matter in 
suspension, the system of training walls would give merely a plus 
result, and their influence be therefore wholly beneficial ; but even 
where there was matter in suspension in the land-locked portion of 
the estuary, — by which I mean that portion between the training 
vrall and the river bank behind it, where alone siltage can go on, 
— and supposing the area of that portion should happen to be very 
small in comparison to the whole extent of the open estuary, the 
plus result would still outweigh the minus , and the balance of the 
influence be for good. What naturally is land-locked might no 
doubt be silted up, and lost to the navigation ; but then, if that 
extent be small compared with what still remains open to the 
entrance of the tides, the river on the whole will be improved. 
If, on the other hand, the land-locked portion of the estuary be 
very large, the total deposit may not only far more than counter- 
balance the additional water gained by extra water passing up the 
improved channel, but leave so little space for the reception of 
tidal water in the estuary, that the amount of scouring water that 
can in future come in from the sea will be insufficient any longer 
