886 BULLETIN OF THE BUEBAXJ OF FISHERIES 



The combined evidence afforded by the drifts from the two lines last discussed 

 points unmistakably to an easterly set as dominating the southern side of Nantucket 

 Sound, with a net movement of the surface water out through the channel between 

 Great Point and Monomoy. The time intervals for the bottles picked up at Rose 

 and Crown Shoal and on the east shore of Nantucket (21 days in each case) show a 

 daily rate of at least IJ^ to 2 miles in this direction at the time. 



With none of the bottles from Nantucket Sound reported within the Gulf of 

 Maine, but abundant evidence of drifts veering to the south and west around Nan- 

 tucket Island and Shoals, it is established with reasonable certainty that the out- 

 flow from Nantucket Sound usually shares in the clockwise eddy movement away 

 from the gulf, which involved the water to the southeast of Cape Cod in 1922 (p. 880) 

 and which is indicated by the measurements made of the currents along the eastern 

 side of Nantucket Shoals (p. 864). 



The fact that three bottles set out in Nantucket Sound in 1924 were picked up in 

 New Jersey, whereas none of the bottles set out abreast the mouth of the sound in 

 1923 were reported so far west, suggests that those that passed eastward out of the 

 sound in 1924 then drifted far enough southward to become involved in the drift 

 followed by the bottles put out on the middle section of the Cape Cod line in the 

 year before. An interesting annual difference thus appears in this respect. 



If this general type of cumulation prevails as constantly from year to year and 

 throughout the summer season, as the bottle drifts suggest, it goes far to explain 

 the fact that tropical fishes, planktonic animals, and floating plants (notably gulf 

 weed), which are so commonly swept from the "Gulf Stream" into Vineyard 

 Sound, only exceptionally enter the gulf around Cape Cod. Passing out of 

 Nantucket Sound to the eastward by the same route followed by the drift bottles, 

 their course would then veer to the southward and so away from the gulf, not into 

 the latter. 



An earlier paragraph, the reader will recall, points out that several bottles from 

 the inner (northern) end of line B, set out of Cape Cod in July, 1922, were carried 

 eastward into the Gulf of Maine, though the majority were swept away from the 

 gulf, locating the division between these two circulating movements (p. 882). 



Series G was set out normal to the coast, about midway of Cape Cod, in 

 August, 1923 (p. 875), in the hope of throwing more light on the southern side of the 

 eddying circulation that dominates the surface waters of the Gulf of Maine. Only 

 5 out of the 100 have been recovered, this being the lowest percentage of recoveries 

 for any of the lines. Two of them, put out, respectively, 4 and 6 miles from the 

 land, were picked up at Nauset near by, one within 2 days after it was set adrift. 

 One bottle, set afloat about 20 miles out at sea, was found 2 months later (October 

 14) floating on the eastern edge of Georges Bank (fig. 176); one launched 5 miles 

 farther out was reported 5 months later from Tiverton, Digby County, on the Nova 

 Scotian shore of the Bay of Fundy, near its mouth; and a fifth, also from the outer 

 end of the line, picked up in Ireland in September a year later, completes the brief 

 list (p. 875). 



Evidently the outer bottles on this line (but not the inner) took part in a drift 

 of the same sort as carried several bottles, set out southeast of Cape Cod in 1922, 

 across to the eastern part of Georges Bank, to the Bay of Fundy, and to France 



