890 BULLETIN OF THE BUREAU OP FISHERIES 



periodic variation, with the dominant movement following around the coast line of 

 the bay in some summers and passing it as a sort of back water at other times. It 

 was in the hope of throwing further light on this secular alternation, especially in its 

 bearing on the involuntary migrations of fish eggs and larvse, that series I and K 

 were set out in the bay in February and May, 1925, and series L, M, and N in April, 

 1926 (p. 877). 



Twenty-three (26 per cent) of the February series of 90 bottles have been recov- 

 ered. Recoveries from bottles set out off the Plymouth shore were distributed as 

 follows: One (No. 74) fromStellwagenBank, 28 miles off Gloucester; one from an equal 

 distance out in the basin of the gulf (fig. 177) ; two from Nantucket; one from the Nova 

 Scotian shore of the Bay of Fundy;'" and one, put out close to the tip of Cape Cod 

 (No. 22), went to Fire Island, New York. 



These drifts, combined, show a definite surface set out of the southern side of 

 the bay, dividing off Cape Cod, where some bottles took the southern route down 

 past Nantucket, and so westward (which so many bottles from the Cape Cod line 

 (line B) followed in July, 1922), while one, at least, was caught up in the southern 

 side of the Gulf of Maine eddy, reproducing the drifts of bottles from the Cape Ann 

 line of 1923 (p. 887). 



The bottles set out in the eastern side of Cape Cod Bay followed a surprisingly 

 definite set eastward and toward Provincetown, no less than 16 out of 21 stranding 

 in that harbor or near by (all of them to the east and most of them well to the north 

 of where they were set adrift) after intervals of 5 to 17 days (usually 5 or 6). Drifts 

 of this sort suggest an anticlockwise movement of the surface water around Cape 

 Cod Bay, with a subsidiary eddy of the same sort in Provincetown Harbor, which 

 finally caught them up as they set northward along the inner shore of the cape. 



Ten bottles set out in Ipswich Bay on April 7 (series J) give definite evidence 

 of a southerly set around Cape Ann and into Massachusetts Bay, one of them having 

 been found at Brant Rock, a few miles north of Plymouth, and two near Race Point, 

 at the tip of Cape Cod, after intervals of 14 to 22 days. A fourth, picked up at 

 Cutler, Me., at the western entrance to the Grand Manan Channel after 106 days, 

 apparently had followed the southern side of the Gulf of Maine eddy, veering south- 

 east, east, and northeast, and so paralleling the drift of bottles set out off Cape Ann 

 in 1923 (line F; p. 887) and at about the same daily rate. A rather definite anti- 

 clockwise drift around the Massachusetts Bay region is thus indicated for winter 

 and early spring by the combined drifts of the February and April series, its southern 

 edge involving Cape Cod Bay but with the water farther north setting more to the 

 eastward and so out past Cape Cod. 



This same type of circulation is still more clearly reflected by the drifts of 40 

 bottles put out in Massachusetts Bay on the 20th to the 22d of that May (series K), 

 drifts so easily interpreted as to demand rather detailed study. Eighteen of these 

 were recovered — the largest percentage (45) for any series yet set out in the Gulf of 

 Maine. 



Following around the bay from north to south we find one or two bottles set 

 out off Manchester" drifting to Marblehead and Nahant, while one bottle set 



" Freeport, Dlgby Cousty. " About 3 miles west of Gloucester. 



