SILTING. 287 



a " close harbour " to encourage silting, inasmuch as the becalmed 

 area under the lee either of a single breakwater projected from 

 the shore, or of an isolated one within the zone of sand-travel 

 even though it be approximately parallel to the coast is often 

 quite as effective in causing deposit to take place. In the latter 

 case the shore being sheltered from the waves, which un- 

 doubtedly play the principal part in the travel of sand and other 

 material along a coast there is, in the absence of a sufficiently 

 strong current, no power by which the sand can be passed on 

 when once it has reached the becalmed area ; the inevitable 

 result being the formation of shoals. 



A good example of this action is to be found in the familiar 

 spit of sand or shingle which 

 so often connects outlying ^^^^^ 

 rocks with the shore, as re- 

 presented by Fig. 97. 



Where littoral currents 

 exist, they may, in a more or 

 less perfect manner, accord- 

 ing to their strength, counter- 

 act the power which the 



waves exert in causing sand to gather in such positions; but 

 littoral currents are often as fickle as the weather, and require 

 very careful study before reliance can be placed upon them. 



It would, therefore, in many cases be dangerous to place a 

 work of magnitude so close to the shore as to shelter it, and it 

 follows that the larger the proposed work the further ought it 

 to be placed from the shore. 



One has not far to look for examples of harbours which 

 have become useless, or dependent upon dredging to an un- 

 necessary extent, by reason of littoral drift having been insuffi- 

 ciently considered, or not properly understood, at the time the 

 designs for them were framed. 



The direct part which wind plays in causing sand to travel 

 along a coast has already been referred to (pp. 20 et seq.). 



Before works are undertaken upon a shore where sand or 

 shingle travels, the physical conditions of the site should be 

 very carefully studied. The investigation should embrace the 

 coast for a long distance on either side of the site of the proposed 

 work, and if the natural form or condition of any point of the 

 coast should be such as to admit of comparison with the proposed 



