906 
BULLETIN OP THE BUREAU OF FISHERIES 
The tracks of three bottles from the mid section of line D, which were picked 
up at the eastern entrance to Frenchmans Bay, and one other that went to the 
vicinity of Petit Manan, are more puzzling. Ostensibly these point to short 
easterly drifts of 8 to 12 miles, and the time intervals are so uniform (33 to 38 
days) 80 that all of them seem to have followed approximately the same route, 
though set out some miles apart. However, the time between release and recovery 
is so long for direct journeys so short, when contrasted with the rapidity with which 
other bottles set out near them drifted in the opposite direction, that it seems 
virtually certain that they followed a roundabout route. Judging from the facts 
that many more bottles stranded to the westward and that all of this series (D) were 
set out on the ebb, it is probable that the four bottles in question also drifted 
westward at first. Their most likely route would then be into Blue Hill Bay with 
the next flood, around Mount Desert Island, and so out again through Frenchmans 
Bay, to strand about Schoodic Promontory and to the eastward of it. Such a drift 
would be consistent with the clockwise circulation to be expected around Mount 
Desert Island, on theoretic grounds (p. 970). In short, the bottles set out off 
Mount Desert in 1923 afford definite proof of a set westward along the coast of 
Maine but no clear evidence of any longshore set in the opposite direction. 
On the basis of the foregoing analysis, the most reasonable explanation of the 
localities where bottles from the Mount Desert, Cape Elizabeth, Cape Ann, and Cape 
Cod series of 1923 were recovered, and of the periods of time between the dates 
they were set afloat and later were picked up, is that bottles from all three lines 
moved in tracks eddying counterclockwise through southwest, through east, to north, 
and veering on successively shorter and shorter radii of curvature. Thus, the few 
bottles from the two southernmost lines, which were found on the Nova Scotian 
coast, probably traveled easterly from the time they were set out (southeast at first, 
then east and northeast) , but the farther north and east along the coast bottles were 
put out, the more they tended to circle to the right of a direct course. It is also 
likely that while the breadth of the track covered by all the bottles in the western 
side of the gulf was something like 100 miles, they tended to converge into a narrower 
track as they approached the eastern side of the gulf. 
In August, September, and October of 1922 and 1923 the center of this eddylike 
circulation seems to have been situated 40 to 60 miles south of Mount Desert 
Island, over the northeastern extension of the deep trough of the gulf. 
The fact that the great majority of the recoveries from Nova Scotia and from 
the Bay of Fundy were from a rather short stretch of coast leads to the conclusion 
that no matter on which line the bottles in question were released, all those that 
drifted across the gulf finally came within the influence of the same south-north 
current, hugging close to the eastern shore. On no other assumption, I believe, is it 
possible to reconcile the facts just stated with the time element (p. 904) and with the 
current measurements that have been taken in that side of the gulf (p. 861). 
The recoveries on the coast of Maine already discussed point to a division of 
this northerly set before it reaches the Bay of Fundy, the greater volume entering 
the bay along its southern shore, but offshoots (which may be only intermittent) 
eo No. 1611 was picked up in Winter Harbor 11 months later, a period so long that there is no way of estimating how far it 
may have traveled en route, or how long it may have lain on the strand. 
