(e) Cross waves in the constant- depth section and transverse waves 

 can be avoided by careful selection of wave period and water depth for 

 each tank width (Barnard and Pritchard, 1972; Madsen, 1974). 



(f) Secondary waves in the constant depth section can be eliminated 

 by programing the generator motion with elliptic functions or by the use 

 of sills placed at the proper location along the tank for each wave 

 period (Hulsbergen, 1974) . 



(g) Variability in profile reflectivity, generation of secondary 

 waves over a shelf, and generation of cross waves over profile segments 

 are phenomena which cannot be avoided or eliminated, but the experi- 

 menters should be aware of the potential of these phenomena to affect 

 profile development. 



(h) As a minimum the experimental conditions discussed in this series 

 of reports should be documented in each movable-bed coastal engineering 

 experiment and model study. 



3. Future Investigation . 



(a) The hypotheses on sources of profile reflectivity variability 

 should be examined one-by-one in fixed-bed experiments. 



(b) More research is needed to quantify the effect of the initial 

 profile slope on the final profile shape. 



(c) More research is needed on how wide a tank must be to assure 

 that the tank walls do not affect a significant part of the profile. 



(d) More basic research is needed on the effect of water tempera- 

 ture on sediment transport in oscillatory flow. 



24 



