880 BULLETIN OF THE BUREAU OF FISHEEIES 



4 months. The line may be divided into three sections, according to the locaHties 

 of recovery: First, an inner section, from Cape Cod across the mouth of Nantucket 

 Sound and skirting the easterly edge of Nantucket Shoals; second, a middle section, 

 from the shoals out nearly to the edge of the continent; and third, the outer end of 

 the line to the seaward of the continental edge. 



Ten bottles out of the 250 set out along the inner section were picked up to the 

 eastward, three of them on the Nova Scotian shore of the Bay of Fundy, one on the 

 northeastern part of Georges Bank, and five (after short drifts) in the south channel 

 and along the northwestern side of Georges Bank (fig. 174). '^^ This last group of 

 recoveries is especially instructive as evidence that the surface water to the south 

 and southeast of Cape Cod was setting in a southeasterly direction at the time. 

 Bottle No. 362, picked up 40 miles to the southeast of the place of its release, after 



5 days' drift, and Nos. 396 and 405, found 30 mUes away after 8 days, can hardly 

 have diverged from a direct line except to follow the spiral tracks induced by the 

 veering tidal currents of this region, unless the dominant set was more rapid at the 

 time than other experiences in the gulf would suggest. *' A southeasterly set is also 

 indicated in this general region by the current measurements carried out by the United 

 States Coast and Geodetic Survey (p. 864). 



The uniformity of these southeasterly drifts makes it likely that the bottles that 

 went from the inner end of line B to the eastern end of Georges Bank and to Nova 

 Scotia also drifted in a southeasterly direction at first, veering to the eastward — i. e., 

 anticlockwise. 



It seems that this inner section of line B followed the boundary of demarkation 

 between this southeasterly set and another drift dh-ected more to the southward from 

 the mouth of Nantucket Sound, veering westward past Nantucket Shoals, because 

 20 bottles from this section were picked up along the southern coast of New England. 

 The fact that current measurements show a general southeasterly set over Nantucket 

 Shoals and a summer set to the west and northwest at the lightship a few miles farther 

 south, makes it more likely that these bottles rounded the shoals than that they 

 crossed the latter. 



It is a question of considerable interest whether 11 bottles, spaced across the 

 eastern entrance to Nantucket Sound, which were picked up along the south shores 

 of Nantucket, Marthas Vineyard, and of New England between Buzzards Bay and 

 and Block Island, drifted directly westward through Nantucket and Vineyard Sounds 

 or whether they also traveled southward around Nantucket Island and Shoals. Of 

 course, a positive answer can not be given; but it seems hardly conceivable that some 

 of them would not have been picked up afloat in the sounds or stranded along shore 

 there if they had gone through, because these beaches are thronged with vacationists. 

 Actually, however, not one of the bottles from line B was found along the northern 

 coast of the sounds, and only one of them on the northern shore of Nantucket, 159 

 days after it was set afloat. One, however, after 30 days afloat, was found 1 mile 

 inside Gay Head at the western end of Marthas Vineyard, where many species of 

 tropical fishes have been recorded in summer. Thus, it seems almost certain that 



«' One bottle from this section went to France. 



" Bottle No. 510 was reported on the northwest slope of Georges Bank, 50 miles from where it was set out, within 3 days . 

 This ostensible drift is so rapid, however, that some error in the reported localit:{ seems probable. 



