910 BULLETIN or THE BUREAU OF FISHEEIES 



One other bottle is recorded by the United States Tlydrographic Office (Pilot 

 Chart for May, 1923; reverse No. 26) as showing a similar drift into the eastern side 

 of the Gulf of Maine from its release, 34 miles south of Cape Sable, September 21, 

 1902, to its recovery near Yarmouth, Nova Scotia, 30 days later. 



CIRCULATION OF THE SUPERFICIAL STRATUM AS INDICATED BY 



SALINITY 



The distribution of sahnity affords a valuable check on the correctness of the 

 circulatory system of the surface stratum, deducible from the drift-bottle experiments 

 and from current measurements. The physical state of the water, together with 

 the horizontal and vertical distribution of density, is the only clue yet available to 

 the nontidal circulation in the deep strata of the gulf. 



The reader will find frequent references to this phase of the subject in the sec- 

 tion devoted to the salinity (p. 701). The distribution of salinity, as a reflection of 

 the circulation of the gulf,| has also been discussed in such detail in earlier reports 

 on the Gulf of Maine explorations (Bigelow, 1914 to 1922) that a brief statement 

 will suffice here. 



With the oceanic water outside the edge of the continent nauch salter than the 

 water in over the banks or alongshore (a rule prevailing all along eastern North 

 America from Florida to the Grand Banks) a high salinity becomes an excellent 

 indicator of any indraft from offshore. On the other hand, the lines of dispersal for 

 land water are to be learned from the distribution of the least sahne water. In the 

 Gulf of Maine the flow of the Nova Scotian current past Cape Sable also tends to 

 freshen the surface wherever its influence reaches. 



Our first summer's cruise (in 1912) was enough to show what subsequent 

 cruises have corroborated, that the freshest water is not localized off the mouths of 

 the several large rivers, as would be the case if the discharges from these simply 

 fanned out, but that it takes the form of a continuous and comparatively narrow belt 

 skirting the coast line. The region where this freshest water does spread farthest 

 out to sea (off Cape Ann and Massachusetts Bay) is some distance southward from 

 the mouth of the Merrimac, the nearest of the large rivers. No fan of low 

 salinity has ever been demonstrated off the mouth of the Kennebec. 



The absence of such a fan off the mouth of any given river may or may not 

 prove the failure of its discharge to drift out to sea, depending on the balance between 

 the activity with which the tides mix the deep with the surface strata there and 

 the volume of fresh water discharged. The river water that runs into the northern side 

 of the gulf, and especially into the Bay of Fundy, is rapidly consumed in this way. 

 Nevertheless, even where mixing is most active, areas of relatively lower salinity off 

 the river mouths might be expected to alternate with areas relatively higher in 

 sahnity along the coast sectors between them, unless some dominant drift in one 

 direction or the other disturbed this idealized picture. When we recall how great 

 a volume of fresh water actually pours into the Gulf of Maine every year (p. 837) it is 

 hardly conceivable that it would exert its chief freshening effect on so narrow a 

 coastwise belt, unless the surface water tended to drift parallel to the land in the one 

 direction or the other. 



