56 



research effort can be optimized. Additionally, such a 

 comparison might lead to directed improvement of these 

 methods to increase their effectiveness as a survey tool. 

 Due to the necessity of avoiding obstacles with the 

 camera-sled, it would be suspected that the camera-sled 

 tends to underestimate coral abundances. The proximity of 

 the camera to the bottom and its angle of incidence with 

 respect to the bottom could also produce different results. 

 The external camera on ALVIN is mounted farther above the 

 bottom and is positioned at a higher angle of incidence 

 than the camera on "Cheep Tow". As a result the camera on 

 "Cheep Tow" views a larger area of the bottom per frame. 

 However, this configuration causes the amplification of 

 errors in estimates of area viewed when the camera's position 

 deviates from horizontal with respect to the bottom. Addition- 

 ally, some of the smaller taxa are more readily seen in the 

 foreground of "Cheep Tow" photographs because of their closer 

 proximity to the camera. Another factor might be differences 

 in the response of motile spibenthos to the two systems. One 

 bias that should not be ignored is the relative difference 

 with regard to randomness between the two techniques. The 

 camera-sled is operated from a surface vessel with little 

 informational feed-back about the area being surveyed. The 

 submersible, on the other hand, is operated under in situ 

 conditions and may reflect biases created by the investigators 



