The rate of the day to day modification differed among the plots . 

 The ripples at the channel mouth site were noticeably subdued after 

 2 days; the intervals after which ripples were first noticed to be 

 subdued in the other sites were: the sand strip, 15 days; the algae 

 flat, 22 days; the patch reef, 9 days. The ripple plot was partly 

 obliterated in the channel mouth after 7 days, in the algae flat, 

 after 30 days, in the patch reef, after 22 days. The ripples on 

 the sand strip were never only partly obliterated. The point at 

 which the ripple pattern completely disappeared was channel mouth, 

 15 days; algae flat and patch reef, 30-35 days; and sand strip, 24 

 days. The ripples clearly were modified and obliterated more 

 rapidly at the channel mouth than at the other sites. 



The style of modification also differed. At the channel mouth site, 

 the ripples were destroyed by the construction of mounds and craters 

 formed by burrowing organisms within the sand. On the sand strip 

 and in the patch reef, the ripples gradually lost their relief and 

 became discernible only by linear aggregations of algal debris that 

 had collected earlier in the ripple troughs. On the algal flat, 

 where the ripples remained sharply defined for a longer period of 

 time, they were destroyed largely by unidentified organisms that 

 dug small holes in the sand. 



The style of disruption of internal structures as shown in box 

 cores taken through the artificially layered sand also differed 

 among the experiment sites. At the channel mouth site the la- 

 minations were largely destroyed by small isolated vertical 

 burrows . At the other sites , a general mixing of the sand ob- 

 literated the layering. This mixing, which was most intense 

 near the sediment-water interface and decreased with depth, pro- 

 bably was caused by organisms crawling across the bottom or 

 feeding in the uppermost part of the sediment. 



The rate of disruption of the layering depended both on the depth 

 of the layer beneath the surface and the style of reworking. In 

 all plots, the lower layer of dyed sand tended to be preserved 

 longer than the upper layer. At the channel mouth site, both 

 layers tended to persist longer than they did at the other sites. 

 This difference in disruption rate resulted from the nature of 

 the mixing and from the fact that sediment cast upon the sand 

 surface by burrowing organisms tended to bury and therefore pro- 

 tect the layering from the activity of organisms that disturbed 

 the uppermost part of the sediment. 



The rate of general sediment surface modification as indicated 

 in daily photographs of the bottom also differed between the 

 canyon mouth and the other sites. The rapid construction of 



VI-27 



