profile, which had a thick iso-velocity layer at the channel axis, 

 may have been caused by a reverse penetration, but the point could 

 not be demonstrated. 



A final question to be asked is, "How widespread is this dis- 

 turbance to the sound channel?" Shallow ridge crests and com- 

 parably saline core water can be foimd up and down the ridge, 

 both north and south of the Azores, so that similar intrusions 

 could occur. The proper approach to answer the question wo\ild be 

 to investigate the salinity of the core layer across the ridge at 

 various latitudes, but to do this from hydrocast data (sac^jles 

 typically spaced two hundred meters apart at the critical depths) 

 is virtually impossible. As a substitute, until continuous salin- 

 ity profiles are the rule, one can trace the salinity along an 

 appropriate depth from a long section across the ridge. The salin- 

 ity will vary with both diffusion and tilting of the isopycnal siir- 

 faces, but these should resiilt in continuous changes with longi- 

 tude, unless a phenomenon such as has been documented here is oc- 

 curring. Depending then on the amount of scatter in the data due 

 to temporal variations, some conjectures might be possible and two 

 such plots are sho>m in Figiire 9* The data is from the Atlantic 

 Ocean Atlas (Fuglister, 196O) and a reference depth of 1,000 meters 

 was selected because it was a frequent saji5)le depth near the depth 

 of maximum salinity. The plot at kO°N is essentially the same area 

 as was studied here, and the expected discontinuity in salinity at 

 the ridge axis is obviously present, agreeing even n magnitude with 

 the data presented above. At 36°N, which runs just south of the 

 Azores, there is some indication of a break at the Mid-Atlantic 

 Ridge crest and a clear discontinuity at about the Azores -Gibraltar 

 Lineament. The latter is deep (3,000 m) at this latitude but rises 

 immediately to the northwest, where it becomes the Azores Plat- 

 form. This type of evidence can be suggestive in locating other pos- 

 sible transition zones. 



Acknowledgements ; 



The observations reported were made during the Mid-Atlantic 

 Ridge survey of cruise No. 82 of R/V CHAIU. Data of this sort is 

 very much a joint effort of the many hands on board, but I especi- 

 ally appreciated the presence of Joseph Phillips and Elizabeth 

 Bunce who provided invaluable assistance by their knowledge of the 

 topography of the ridge. Also making notabee contributions were 

 William Rossow, who concluded the lowering program after I had left 

 the cruise, and Mrs. Maxine Jones who programmed the date reduction. 



The work was performed under contract Nonr-i«D29(00) with the 

 Office of Naval Research. 



419 



