491) 



John B. Tait 



support of the suggestion is contained in another aspect of the results of this com- 

 prehensive investigation of the hydrography of the Faroe-Shetland Channel which is 

 about to appear elsewhere* in extenso. 



II was mentioned earlier that the remarkably close agreements recorded between 

 the contemporary paired sections across the Channel depended also in a number of 

 cases and as it turned out depended very materially in some instances, on a certain 

 condition which referred to both sections. This condition was the re-inclusion of 

 apparently aberrant original observations of temperature or sahmty, principally 



FAROE BANK,- BUTT OF LEWIS 



f R.S "EXPLORER" 

 20-22 JUNE I9J3 



Fig. 3 



the latter, which by former standards were customarily excluded, or replaced by 

 interpolation from smooth curves, if they implied more or less radical inversion of 

 density in the vertical water column. In the present investigation, the years in which 

 this condition of reinsertion of such observations had most frequently to be applied 

 were those between 1931 and 1936 inclusive, when considerable numbers of high and 

 even of extraordinarily high salinity values were registered, of maximum magnitude 

 and frequency in the year 1933 in the Faroe-Shetland Channel, in 1934 in the northern 

 North Sea, and almost without exception as subsurface values. It was in the latter 

 region, in fact, that these recurring abnormally high saMnity records were first 



* In the Scottish Home Department's series of " Marine Research " publications. 



