form of the wave crest in the lee of breakwater is distorted again by a short, 

 almost straight portion of orest, connecting the more adx r anced circular crast in 

 deep«*water and the inoident wave in shallow water. The change from a straight 

 orest to a ourved one appears to be more abrupt than in Run 8, However, the 

 diffraotion pattern depends greatly upon the ratio of water depths in the lee 

 of breakwater to the depths outside. For the uniform depth of water it would 

 be expected that the wave crests in the shadow of the break-water would assume 

 an almost circular form, with their oenter at the tip of breakwater. As for 

 the heights of dif fraoted waves - shallow water in the lee results in higher 

 diffracted waves than if deep water exists in that area. 



Some breakwaters have gaps through which vessels may enter sheltered waters* 

 "When waves pass through a gap, diffraction occurs at the two ends of the gap. 

 In the case of a relatively narrow gap (compared with the wave length), the 

 diffraction of waves in the lee of breakwater will be different than that 

 around a semi-infinite breakwater tip. Theories have been developed for this 

 oondition (Ref. 3 and 10) and generalized diagrams developed by Johnson (6) 

 for various conditions of wave approach and wave length. These diagrams can 

 be used as transparent overlays, and by moving them over a chart of a harbor,- 

 the location of the breakwater to give the most desirable protection can be 

 obtained. 



It has been found that when the gap width is in excess of about five wave 

 lengths, the diffraotion patterns at eaoh side of the opening are nearly inde- 

 pendent of eaoh other. In such cases the pattern given for a semi-infinite 

 breakwater oan be used to estimate the height and direction of waves on the 

 leeward side. As a comparison, it might be mentioned here, that this fact seems 

 to be valid also for the diffraotion pattern around the ends of a finite-length 

 breakwater, gjlhis has been demonstrated in Runs 15 to 25 (Figures 3f and 3g) 

 wherein waves /diffracted around a vertical walled cylinder. When the diameter 

 of the oylinder approaohed 6 wave lengths, the diffraction patterns around eaoh 

 of the ends appear to be nearly independent as far as the alinement of the wave 

 crests is oonoerned. This fact is discussed further in the seotion on "Islands"* 



Run 10 (Figure 3b) was made to demonstrate diffraotion at a breakwater gap. 

 A breakwater was introduced which extended completely aoross the tank, with an 

 opening of about 1.2 wave lengths wide in its oenter. The water depth was uni- 

 form and the wave period was T s 0.22 sec. The nearly circular wave pattern, 

 with oenter in the middle of the gap, oan be observed in the photograph. The 

 change of wave heights is indicated by the intensity of white crest-lines. 

 The waves were relatively high close to the gap and along a line parallel to the 

 direction of wave approaoh and passing through the oenter of the gap. The wave 

 height deoreased rapidly within the shadow of the breakwater in the direction 

 perpendioular to the direction of wave advance, and less rapidly in direction 

 of wave approach. 



It has been shown by Putnam and Arthur (Ref. 2) that the geometrical shape of 

 the breakwater tip, within the limits of their investigation, had no material 

 influence on the wave diffraotion pattern. With a sharp corner, it was 

 observed that a seoondary disturbance originated at the corners with propagated 

 capillary waves with a oiroular pattern superposed upon the main gravity-wave 

 system. This is very clearly demonstrated in Run 10 (Figure 3b). A better 

 oonoeption of this phenomenon oan be obtained by observing the movies of this 

 run< however, the pattern oan be distinguished also in the photo Bhown in 



8 



