NO. 5 BARNARD: AMPHIPODA 31 



vals. Thus, benthic currents as well as areas of upvvelling might be de- 

 tected simultaneously. Epibenthic current meters would have to be paired 

 to detect both horizontal and vertical motion. 



5. A series of sediment traps at substrate level and above should be 

 used to collect sediments for studies of depositional rates and accretion 

 of organic matter. Traps might be attached to long flexible arms so that 

 they could be positioned remotely adjacent to canyon walls or at the 

 bases of declivities, on benches and in "plunge pools." Large deep tubs 

 of sediments devoid of metazoan life might be established and monitored 

 for biotic succession. The seasonal variability and origin of settling lar- 

 vae could be determined simultaneously. A series of standards might be 

 developed that would indicate the stage of development of a specific 

 sample, thus reflecting the temporal aspects of any previous environ- 

 mental catastrophe. 



6. Identification of the kinds and sources of organic debris. 



7. Determination of the viability of those species that may be exist- 

 ing vegetatively (non- reproducing organisms recruited from shallow 

 water). 



8. Establishment of an alarm system in supposedly active canyons 

 for warning of turbidity flows so that post-catastrophic sampling and 

 exploration by undersea vehicles could be undertaken to monitor the 

 return of the fauna to climax conditions. 



SUMMARY 



1. Soft bottoms in canyon heads, 15-100 m, have a more diverse 

 algal-dwelling amphipod fauna than is found on the coastal shelf in the 

 same depths, but the fauna is not identifiable with that from sand of 

 either very shallow water (4-10 m), the coastal shelf (11-91 m), or the 

 upper coastal slope (92-183 m), because it represents a mixture of ele- 

 ments from those areas. Paraphoxus bicuspidatus, a common inhabitant 

 of shelf and slope, is scarce. 



2. In depths of 101-200 m the benthic amphipod fauna compares 

 favorably with upper coastal slopes of 92-183 m and Paraphoxus bicus- 

 pidatus occurs abundantly. 



3. Although standing crop is erratic from sample to sample, there is 

 a significant decrease between depths of 600 and 700 m from a level of 



